anchor
stringlengths
723
2.15k
positive
stringlengths
153
8.2k
Given the stale inspection and broad query, enumerate records from the PDF source, but only read, and return the metadata and filters. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "e...
Apply this capability when the next decision depends on resource metadata and filters.. Capabilities: search metadata. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: metadata catalog required. Side effects: read. Arguments: filters (object required), sort (st...
We've seen snippets but not full evidence; now resolve contradictions between Excel sources by filtering on metadata, and output the schema fields, types, and structure. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan...
This candidate supports controlled inspection of fields, types, and structure.. Capabilities: inspect structured schema. Accepts: csv, excel, sqlite. Returns: schema. Evidence role: structure discovery. Prerequisites: source selected. Constraints: structured source required. Side effects: read. Arguments: include types...
Don't just read the code or CSV; instead, join matching records from the mixed sources using lexical match, and output the assessment and governance state. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": f...
Use this candidate when the task requires matching records.. Capabilities: search structured records. Accepts: csv, excel, sqlite. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: schema known. Constraints: structured source required. Side effects: read. Arguments: limit (integer require...
To verify the result, read the PDF and text sources, extract document structure (sections and landmarks) and evidence, but do not write anything. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "compatibility"} Agent State: {"phase": "planning", ...
This operation handles sections and structural landmarks while preserving the current task state.. Capabilities: inspect document structure. Accepts: pdf. Returns: document structure. Evidence role: structure discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: lan...
We have 2 steps left; which pages or ranges should I check to resolve the disputed field definitions? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "lookup_semantic"} Agent State: {"p...
Invoke the candidate to determine the relevant pages or ranges from the available resources.. Capabilities: search document pages. Accepts: pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: page range (...
I need to recover from an incomplete attempt; what should I do next to enumerate all records in the selected resource for consistency? Decision: route Plan: {"active": false, "operation": "enumerate_records"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fresh_sufficient"} ...
The operation produces a reliable result for the selected resource under current constraints.. Capabilities: read content. Accepts: text, code. Returns: content. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: local source required. Side effects: read. Arguments: offset (integer optiona...
Given the catalog, which symbols or definitions should I inspect to verify absence and resource limits? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "absence_verification"} Agent Stat...
Apply this capability when the next decision depends on symbols and definitions.. Capabilities: inspect code structure. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Arguments: depth (integer re...
What should I do next to compare candidate results for failure recovery? Decision: route Plan: {"active": true, "operation": "compare"} Agent State: {"phase": "retrieval", "question_length_band": "short", "state_name": "retrieval"} Source State: {"available_modalities": ["csv", "metadata"], "evidence_topology_observed"...
This candidate supports controlled inspection of a candidate result.. Capabilities: inspect evidence. Accepts: evidence candidates. Returns: evidence. Evidence role: evidence inspection. Prerequisites: evidence candidate selected. Constraints: candidate evidence required. Side effects: read. Arguments: claim (string op...
Given the stale access control assessment, which candidate evidence source should I check next to find the latest permissions? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "latest_val...
Use this candidate when the task requires missing surrounding context.. Capabilities: expand context. Accepts: evidence candidates, evidence. Returns: evidence. Evidence role: context expansion. Prerequisites: evidence candidate selected. Constraints: incomplete context required. Side effects: read. Arguments: result i...
I've looked at multiple sources for event delivery, but they don't align. What should I do next? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"action_family":...
This operation handles agreement or conflict between results while preserving the current task state.. Capabilities: compare evidence. Accepts: evidence. Returns: evidence comparison. Evidence role: evidence synthesis. Prerequisites: evidence inspected. Constraints: multiple evidence required. Side effects: none. Argum...
I need to resolve a contradiction in failure recovery requirements. What tool operation should I perform next? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"a...
Invoke the candidate to determine task requirements and coverage from the available resources.. Capabilities: update requirements. Accepts: evidence, agent state. Returns: governance state. Evidence role: governance update. Prerequisites: evidence inspected. Constraints: tracked requirements required. Side effects: loc...
With only one step left and multiple passages to inspect, how can I determine if I have enough information for the next workflow step? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "enumerate_sources"} Agent State: {"phase": "planning", "questi...
The operation produces a reliable result for whether available information is sufficient under current constraints.. Capabilities: assess evidence. Accepts: evidence, governance state. Returns: assessment. Evidence role: governance assessment. Prerequisites: requirements updated. Constraints: canonical evidence set req...
Given the complete requirements and fresh assessment, but with partial evidence inspection and version noise, which operation should I perform next to join the evidence candidates without writing external state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false...
Apply this capability when the next decision depends on the selected result or next action.. Capabilities: finalize selection. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh sufficient assessment required. Side ...
I've already checked some evidence and found a disputed requirement, but there's still one unresolved and a long document to search. What should I do next to gather more information without altering the governance state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"activ...
This candidate supports controlled inspection of the best next information-gathering step.. Capabilities: plan retrieval. Accepts: agent state, text. Returns: plan. Evidence role: none. Prerequisites: objective available. Constraints: none. Side effects: local state write. Arguments: objective (string required), priori...
I have stale assessments and only partial requirements, but I've seen full context across formats. Should I search for more sources or inspect the available resources first, given the side-effect constraint? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_us...
Use this candidate when the task requires the available resources.. Capabilities: list sources. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: none. Side effects: read. Arguments: filters (object optional), scope (string required)
I've inspected multiple sources that contradict each other, and the requirements are complete. How should I verify the result by extracting relevant content from the PDF evidence, without modifying anything? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_us...
This operation handles the relevant content while preserving the current task state.. Capabilities: search content. Accepts: text, pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: indexed sources required. Side effects: read. Arguments: limit (integer o...
With only two steps left, which operation should I prioritize to pin down the exact identifier from the inspected evidence? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "lookup_semantic"} Agent State: {"phase": "retrieval", "question_length_ba...
The operation produces a reliable result for an exact name, code, or token under current constraints.. Capabilities: exact pattern search, identifier search. Accepts: text, code. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: textual source required. Side ...
My previous attempt was incomplete. What should I do next to get the latest resource metadata and filters for the event delivery? Decision: route Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fresh_sufficient"} ...
The operation produces a reliable result for resource metadata and filters under current constraints.. Capabilities: search metadata. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: metadata catalog required. Side effects: read. Arguments: filters (object requ...
I need to know the fields and structure for the telemetry trace. Which catalog operation should I perform next? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"...
Apply this capability when the next decision depends on fields, types, and structure.. Capabilities: inspect structured schema. Accepts: csv, excel, sqlite. Returns: schema. Evidence role: structure discovery. Prerequisites: source selected. Constraints: structured source required. Side effects: read. Arguments: includ...
What should I do next to find the matching records for the resource limit? Decision: route Plan: {"active": true, "operation": "lookup_exact"} Agent State: {"phase": "retrieval", "question_length_band": "short", "state_name": "retrieval"} Source State: {"available_modalities": ["sqlite", "metadata"], "evidence_topology...
This candidate supports controlled inspection of matching records.. Capabilities: search structured records. Accepts: csv, excel, sqlite. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: schema known. Constraints: structured source required. Side effects: read. Arguments: limit (integer ...
Given the schema ambiguity, list the sections and structural landmarks in the PDF source. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "sy...
Use this candidate when the task requires sections and structural landmarks.. Capabilities: inspect document structure. Accepts: pdf. Returns: document structure. Evidence role: structure discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: landmark (string optiona...
Given the split context, find the relevant pages or ranges for telemetry and trace requirements. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"action_family":...
This operation handles the relevant pages or ranges while preserving the current task state.. Capabilities: search document pages. Accepts: pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: page range (...
To resolve the disputed requirement, determine which resource is the latest version. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phase": "go...
Invoke the candidate to determine the selected resource from the available resources.. Capabilities: read content. Accepts: text, code. Returns: content. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: local source required. Side effects: read. Arguments: offset (integer optional), reso...
Given the version noise and remaining step, extract the symbols and definitions from the code source. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"action_fam...
The operation produces a reliable result for symbols and definitions under current constraints.. Capabilities: inspect code structure. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Arguments: de...
Given the complete evidence and unresolved contradictions, should we next validate a candidate result against the version compatibility metadata, ensuring only read operations? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1...
Apply this capability when the next decision depends on a candidate result.. Capabilities: inspect evidence. Accepts: evidence candidates. Returns: evidence. Evidence role: evidence inspection. Prerequisites: evidence candidate selected. Constraints: candidate evidence required. Side effects: read. Arguments: claim (st...
We've only seen one record so far; what should we check next to get the missing surrounding context for the exact identifier, given the version noise? Decision: route Plan: {"active": true, "operation": "enumerate_records"} Agent State: {"phase": "retrieval", "question_length_band": "long", "state_name": "retrieval"} S...
This candidate supports controlled inspection of missing surrounding context.. Capabilities: expand context. Accepts: evidence candidates, evidence. Returns: evidence. Evidence role: context expansion. Prerequisites: evidence candidate selected. Constraints: incomplete context required. Side effects: read. Arguments: r...
Between comparing the two inspected results for agreement versus searching for a new source, which is the right next move for failure recovery, given the stale assessment? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "latest_value_selection"}...
Use this candidate when the task requires agreement or conflict between results.. Capabilities: compare evidence. Accepts: evidence. Returns: evidence comparison. Evidence role: evidence synthesis. Prerequisites: evidence inspected. Constraints: multiple evidence required. Side effects: none. Arguments: comparison (str...
To verify the joined result against access rules, should we next inspect the task requirements coverage by checking the relevant evidence candidates, given the acronym obstacle? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "join"} Agent State...
This operation handles task requirements and coverage while preserving the current task state.. Capabilities: update requirements. Accepts: evidence, agent state. Returns: governance state. Evidence role: governance update. Prerequisites: evidence inspected. Constraints: tracked requirements required. Side effects: loc...
With only 8 steps left, should I first verify whether the existing evidence is enough to settle the disputed access rule, or must I gather more? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use",...
Invoke the candidate to determine whether available information is sufficient from the available resources.. Capabilities: assess evidence. Accepts: evidence, governance state. Returns: assessment. Evidence role: governance assessment. Prerequisites: requirements updated. Constraints: canonical evidence set required. S...
The last attempt was incomplete. Based on the fresh evidence and the 2 remaining steps, what should be the next operation to finish the data transformation? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": ...
The operation produces a reliable result for the selected result or next action under current constraints.. Capabilities: finalize selection. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh sufficient assessment ...
For telemetry and trace requirements, what is the best next step to gather information, given the catalog is only partially known? Decision: route Plan: {"active": true, "operation": "compare"} Agent State: {"phase": "retrieval", "question_length_band": "long", "state_name": "retrieval"} Source State: {"available_modal...
Apply this capability when the next decision depends on the best next information-gathering step.. Capabilities: plan retrieval. Accepts: agent state, text. Returns: plan. Evidence role: none. Prerequisites: objective available. Constraints: none. Side effects: local state write. Arguments: objective (string required),...
What resources are available for version compatibility analysis? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "source_discovery", "question_length_band": "short", "state_name": "source_discovery"} S...
This candidate supports controlled inspection of the available resources.. Capabilities: list sources. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: none. Side effects: read. Arguments: filters (object optional), scope (string required)
Given the SQLite source and schema ambiguity, what should I do next to find the relevant content? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "join"} Agent State: {"phase": "inspect...
Use this candidate when the task requires the relevant content.. Capabilities: search content. Accepts: text, pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: indexed sources required. Side effects: read. Arguments: limit (integer optional), query (stri...
What is the next step to retrieve an exact name or code from the Excel file, considering the schema ambiguity? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "lookup_semantic"} Agent S...
This operation handles an exact name, code, or token while preserving the current task state.. Capabilities: exact pattern search. Accepts: text, code. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: textual source required. Side effects: read. Arguments: c...
I need to enumerate records from the PDF. What should I do next to get the resource metadata and filters? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "enumerate_records"} Agent State: {"phase": "terminal_readiness", "question_length_band": "...
Invoke the candidate to determine resource metadata and filters from the available resources.. Capabilities: search metadata. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: metadata catalog required. Side effects: read. Arguments: filters (object required), s...
Given the planning phase, complete requirements, and known Excel source, what operation should I perform to compare fields, types, and structure while respecting resource limits? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step":...
The operation produces a reliable result for fields, types, and structure under current constraints.. Capabilities: inspect structured schema. Accepts: csv, excel, sqlite. Returns: schema. Evidence role: structure discovery. Prerequisites: source selected. Constraints: structured source required. Side effects: read. Ar...
Given the stale assessment and exact identifier, which read-only operation should I perform next to verify absence of matching records before any state write? Decision: route Plan: {"active": false, "operation": "absence_verification"} Agent State: {"phase": "governance", "question_length_band": "long", "state_name": "...
Apply this capability when the next decision depends on matching records.. Capabilities: search structured records. Accepts: csv, excel, sqlite. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: schema known. Constraints: structured source required. Side effects: read. Arguments: limit (i...
After checking only one passage and finding multiple nearby candidates, what should I do next to locate the correct structural landmark for the latest value? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "latest_value_selection"} Agent State: ...
This candidate supports controlled inspection of sections and structural landmarks.. Capabilities: inspect document structure. Accepts: pdf. Returns: document structure. Evidence role: structure discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: landmark (string ...
Given that we've already inspected multiple sources but the assessment is stale, should I now update the governance state or instead read the relevant pages to find the latest value? Decision: route Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phase": "governance", "question_length_band...
Use this candidate when the task requires the relevant pages or ranges.. Capabilities: search document pages. Accepts: pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: page range (array required), quer...
How can I verify the compatibility of the selected resource using only read-only operations, given contradictory evidence snippets and a known source inventory? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "compatibility"} Agent State: {"phas...
This operation handles the selected resource while preserving the current task state.. Capabilities: read content. Accepts: text, code. Returns: content. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: local source required. Side effects: read. Arguments: offset (integer optional), reso...
Given our limited steps, which next operation should we prioritize to resolve the disputed field definitions? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "enumerate_records"} Agent ...
Invoke the candidate to determine symbols and definitions from the available resources.. Capabilities: inspect code structure. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Arguments: depth (int...
After that incomplete attempt, what should we do next to find the relevant technical instruction? Decision: route Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "inspection", "question_length_band": "short", "state_name": "partial_evidence"} Source State: {"available_modalities": ["csv", "metada...
The operation produces a reliable result for a candidate result under current constraints.. Capabilities: inspect evidence. Accepts: evidence candidates. Returns: evidence. Evidence role: evidence inspection. Prerequisites: evidence candidate selected. Constraints: candidate evidence required. Side effects: read. Argum...
What source or catalog operation should we perform next to get the missing context for the transformation? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"actio...
Apply this capability when the next decision depends on missing surrounding context.. Capabilities: expand context. Accepts: evidence candidates, evidence. Returns: evidence. Evidence role: context expansion. Prerequisites: evidence candidate selected. Constraints: incomplete context required. Side effects: read. Argum...
What should we do next to resolve the contradiction between the event delivery results? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"action_family": "tool_us...
This candidate supports controlled inspection of agreement or conflict between results.. Capabilities: compare evidence. Accepts: evidence. Returns: evidence comparison. Evidence role: evidence synthesis. Prerequisites: evidence inspected. Constraints: multiple evidence required. Side effects: none. Arguments: comparis...
I've found two passages mentioning version numbers but they don't match the entity I need. What should I do next to cover all requirements? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation"...
Use this candidate when the task requires task requirements and coverage.. Capabilities: update requirements. Accepts: evidence, agent state. Returns: governance state. Evidence role: governance update. Prerequisites: evidence inspected. Constraints: tracked requirements required. Side effects: local state write. Argum...
I've inspected multiple sources but still can't determine if the data meets quality standards. Can you help me decide what to do? Decision: route Plan: {"active": true, "operation": "compare"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name": "initial"} Source State: {"available_modalitie...
This operation handles whether available information is sufficient while preserving the current task state.. Capabilities: assess evidence. Accepts: evidence, governance state. Returns: assessment. Evidence role: governance assessment. Prerequisites: requirements updated. Constraints: canonical evidence set required. S...
I need to combine evidence from different passages to check access permissions. What is the next action to achieve that? Decision: route Plan: {"active": false, "operation": "join"} Agent State: {"phase": "inspection", "question_length_band": "long", "state_name": "partial_evidence"} Source State: {"available_modalitie...
The operation produces a reliable result for the selected result or next action under current constraints.. Capabilities: finalize selection, emit terminal result. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh ...
I've searched many times but only have snippets from a PDF. What should I do next to find the exact version info, considering I have limited steps? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_us...
The operation produces a reliable result for the best next information-gathering step under current constraints.. Capabilities: plan retrieval. Accepts: agent state, text. Returns: plan. Evidence role: none. Prerequisites: objective available. Constraints: none. Side effects: local state write. Arguments: objective (st...
I've inspected multiple sources but the assessment is stale and two requirements are disputed. What resources are available to help with field definitions and compatibility? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] ...
Apply this capability when the next decision depends on the available resources.. Capabilities: list sources. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: none. Side effects: read. Arguments: filters (object optional), scope (string required)
We've only seen snippets so far and haven't searched yet. What content is available for understanding access rules and permissions? Decision: route Plan: {"active": true, "operation": "enumerate_sources"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name": "initial"} Source State: {"availab...
This candidate supports controlled inspection of the relevant content.. Capabilities: search content. Accepts: text, pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: indexed sources required. Side effects: read. Arguments: limit (integer optional), quer...
We need to recover from a failure. Should we search for the exact code or token, or instead enumerate all sources? Decision: route Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "governance", "question_length_band": "long", "state_name": "insufficient"} Source State: {"available_modali...
Use this candidate when the task requires an exact name, code, or token.. Capabilities: exact pattern search. Accepts: text, code. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: textual source required. Side effects: read. Arguments: case sensitive (boolea...
After exhausting sources and finding no evidence, how can I verify the next workflow step using resource metadata and filters? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak",...
This operation handles resource metadata and filters while preserving the current task state.. Capabilities: search metadata. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: metadata catalog required. Side effects: read. Arguments: filters (object required), s...
With no steps left and a disputed requirement, what should we prioritize to clarify fields, types, and structure? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, ...
Invoke the candidate to determine fields, types, and structure from the available resources.. Capabilities: inspect structured schema. Accepts: csv, excel, sqlite. Returns: schema. Evidence role: structure discovery. Prerequisites: source selected. Constraints: structured source required. Side effects: read. Arguments:...
The previous attempt was incomplete; what should we do next to match records and advance the workflow? Decision: route Plan: {"active": true, "operation": "contradiction_resolution"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name": "initial"} Source State: {"available_modalities": ["csv"...
The operation produces a reliable result for matching records under current constraints.. Capabilities: search structured records. Accepts: csv, excel, sqlite. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: schema known. Constraints: structured source required. Side effects: read. Argu...
We need to enumerate records for telemetry; what resource should we consult to find sections and structural landmarks? Decision: route Plan: {"active": false, "operation": "enumerate_records"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fresh_sufficient"} Source State: {"...
Apply this capability when the next decision depends on sections and structural landmarks.. Capabilities: inspect document structure. Accepts: pdf. Returns: document structure. Evidence role: structure discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: landmark (...
What is the next action to resolve the contradiction by locating the relevant pages or ranges? Decision: route Plan: {"active": false, "operation": "contradiction_resolution"} Agent State: {"phase": "source_discovery", "question_length_band": "short", "state_name": "source_discovery"} Source State: {"available_modaliti...
This candidate supports controlled inspection of the relevant pages or ranges.. Capabilities: search document pages. Accepts: pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: page range (array required...
Which resource should be used to resolve the contradiction in the data transformation, given the stale assessment and the need for a metadata filter? Decision: route Plan: {"active": false, "operation": "contradiction_resolution"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name"...
Use this candidate when the task requires the selected resource.. Capabilities: read content. Accepts: text, code. Returns: content. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: local source required. Side effects: read. Arguments: offset (integer optional), resource (string required...
What should be inspected next to locate the exact symbols and definitions for resource limits and lifecycle, without naming the source? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result"...
This operation handles symbols and definitions while preserving the current task state.. Capabilities: inspect code structure. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Arguments: depth (int...
To complete the data transformation, which candidate result should be selected as evidence, given the broad query and the need to filter by metadata? Decision: route Plan: {"active": false, "operation": "lookup_semantic"} Agent State: {"phase": "synthesis", "question_length_band": "long", "state_name": "contradiction"}...
Invoke the candidate to determine a candidate result from the available resources.. Capabilities: inspect evidence. Accepts: evidence candidates. Returns: evidence. Evidence role: evidence inspection. Prerequisites: evidence candidate selected. Constraints: candidate evidence required. Side effects: read. Arguments: cl...
Given the need for consistency across sources and the missing context, what should be compared to resolve the acronym issue, considering the governance state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family...
The operation produces a reliable result for missing surrounding context under current constraints.. Capabilities: expand context. Accepts: evidence candidates, evidence. Returns: evidence. Evidence role: context expansion. Prerequisites: evidence candidate selected. Constraints: incomplete context required. Side effec...
Given the contradictory evidence from multiple sources, how should I proceed to verify the absence of the item without modifying any data? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "resu...
Apply this capability when the next decision depends on agreement or conflict between results.. Capabilities: compare evidence. Accepts: evidence. Returns: evidence comparison. Evidence role: evidence synthesis. Prerequisites: evidence inspected. Constraints: multiple evidence required. Side effects: none. Arguments: c...
I've checked the relevant documents already; what remains to be done to cover all task requirements? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"action_fami...
This candidate supports controlled inspection of task requirements and coverage.. Capabilities: update requirements. Accepts: evidence, agent state. Returns: governance state. Evidence role: governance update. Prerequisites: evidence inspected. Constraints: tracked requirements required. Side effects: local state write...
Should I extract the exact identifier from the PDF or is the current evidence sufficient to proceed with the transformation? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "...
This operation handles whether available information is sufficient while preserving the current task state.. Capabilities: assess evidence, validate sufficiency. Accepts: evidence, governance state. Returns: assessment. Evidence role: governance assessment. Prerequisites: requirements updated. Constraints: canonical ev...
How can I verify that the next workflow step has been executed correctly, using the available evidence? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "compatibility"} Agent State: {"p...
This operation handles the selected result or next action while preserving the current task state.. Capabilities: finalize selection. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh sufficient assessment required...
Given the remaining steps are zero and the requirement is disputed, what is the highest-priority information-gathering action to resolve the conflict? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool...
Invoke the candidate to determine the best next information-gathering step from the available resources.. Capabilities: plan retrieval. Accepts: agent state, text. Returns: plan. Evidence role: none. Prerequisites: objective available. Constraints: none. Side effects: local state write. Arguments: objective (string req...
The previous attempt was incomplete. What resources are available to enumerate records for the disputed field definitions? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "enumerate_records"} Agent State: {"phase": "governance", "question_length...
The operation produces a reliable result for the available resources under current constraints.. Capabilities: list sources. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: none. Side effects: read. Arguments: filters (object optional), scope (string required)
What content is relevant for comparing against the schema-bound requirement in the failure recovery context? Decision: route Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fresh_sufficient"} Source State: {"available_modalities...
Apply this capability when the next decision depends on the relevant content.. Capabilities: search content. Accepts: text, pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: indexed sources required. Side effects: read. Arguments: limit (integer optional...
What is the exact name, code, or token needed to verify absence for the required data transformation? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "absence_verification"} Agent State...
This candidate supports controlled inspection of an exact name, code, or token.. Capabilities: exact pattern search. Accepts: text, code. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: textual source required. Side effects: read. Arguments: case sensitive ...
I have a near-duplicate issue across sources. What should I do next to compare them? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "inspection", "question_length_band": "short", "state_name": "partial_evidence...
This operation handles resource metadata and filters while preserving the current task state.. Capabilities: search metadata, filter sources. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: metadata catalog required. Side effects: read. Arguments: filters (obj...
What fields and types should I check next to validate the CSV data? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "terminal_readiness", "question_length_band": "short", "state_name": "fresh_sufficien...
This operation handles fields, types, and structure while preserving the current task state.. Capabilities: inspect structured schema. Accepts: csv, excel, sqlite. Returns: schema. Evidence role: structure discovery. Prerequisites: source selected. Constraints: structured source required. Side effects: read. Arguments:...
I need to transform records but have identifier variations. What should I do next? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "source_discovery", ...
Invoke the candidate to determine matching records from the available resources.. Capabilities: search structured records. Accepts: csv, excel, sqlite. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: schema known. Constraints: structured source required. Side effects: read. Arguments: l...
For the disputed requirement, what structural landmarks should I locate next in the document? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "compatibility"} Agent State: {"phase": "source_discovery", "question_length_band": "short", "state_nam...
The operation produces a reliable result for sections and structural landmarks under current constraints.. Capabilities: inspect document structure. Accepts: pdf. Returns: document structure. Evidence role: structure discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Argume...
Given the stale assessment and partial source inventory, find the relevant pages or ranges for access rules without modifying local state. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "resu...
Apply this capability when the next decision depends on the relevant pages or ranges.. Capabilities: search document pages. Accepts: pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: page range (array r...
We've checked some snippets but not the full source; identify the selected resource to resolve the contradiction, using only reads and no state changes. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "to...
This candidate supports controlled inspection of the selected resource.. Capabilities: read content. Accepts: text, code. Returns: content. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: local source required. Side effects: read. Arguments: offset (integer optional), resource (string r...
We need the latest symbol definitions, but don't just fetch the document structure—extract the specific symbols and definitions for event retry behavior. Decision: route Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_nam...
Use this candidate when the task requires symbols and definitions.. Capabilities: inspect code structure. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Arguments: depth (integer required), resou...
How can we verify the candidate result for version compatibility by examining evidence, without altering any state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "lookup_semantic"} Agent State: {"phase": "inspection", "question_length_band": "...
This operation handles a candidate result while preserving the current task state.. Capabilities: inspect evidence. Accepts: evidence candidates. Returns: evidence. Evidence role: evidence inspection. Prerequisites: evidence candidate selected. Constraints: candidate evidence required. Side effects: read. Arguments: cl...
With no steps left, prioritize which source to inspect next for missing context, given partial evidence and stale assessment. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "enumerate_...
Invoke the candidate to determine missing surrounding context from the available resources.. Capabilities: expand context. Accepts: evidence candidates, evidence. Returns: evidence. Evidence role: context expansion. Prerequisites: evidence candidate selected. Constraints: incomplete context required. Side effects: read...
After an ambiguous attempt, which next action resolves the contradiction between results, given fresh assessment and one evidence passage? Decision: route Plan: {"active": false, "operation": "contradiction_resolution"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fresh_su...
The operation produces a reliable result for agreement or conflict between results under current constraints.. Capabilities: compare evidence. Accepts: evidence. Returns: evidence comparison. Evidence role: evidence synthesis. Prerequisites: evidence inspected. Constraints: multiple evidence required. Side effects: non...
Which resource should be consulted to cover the relevant technical instruction, given partial inventory and snippets only? Decision: route Plan: {"active": false, "operation": "contradiction_resolution"} Agent State: {"phase": "inspection", "question_length_band": "long", "state_name": "partial_evidence"} Source State:...
This candidate supports controlled inspection of task requirements and coverage.. Capabilities: update requirements, track coverage. Accepts: evidence, agent state. Returns: governance state. Evidence role: governance update. Prerequisites: evidence inspected. Constraints: tracked requirements required. Side effects: l...
What is the next action to determine if available information suffices, given unknown inventory and multiple inspected sources? Decision: route Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "governance", "question_length_band": "long", "state_name": "insufficient"} Source State: {"ava...
This candidate supports controlled inspection of whether available information is sufficient.. Capabilities: assess evidence. Accepts: evidence, governance state. Returns: assessment. Evidence role: governance assessment. Prerequisites: requirements updated. Constraints: canonical evidence set required. Side effects: l...
Check if the current version is compatible with the target system. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "compatibility"} Agent State: {"phase": "governance", "question_length...
Use this candidate when the task requires the selected result or next action.. Capabilities: finalize selection. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh sufficient assessment required. Side effects: local...
We have conflicting requirements; what's the best next move to clarify access permissions? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "compatibility"} Agent State: {"phase": "retrieval", "question_length_band": "short", "state_name": "retrie...
This operation handles the best next information-gathering step while preserving the current task state.. Capabilities: plan retrieval. Accepts: agent state, text. Returns: plan. Evidence role: none. Prerequisites: objective available. Constraints: none. Side effects: local state write. Arguments: objective (string req...
We need to ensure consistency across sources; what resources are available to help with that? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "join"} Agent State: {"phase": "source_discovery", "question_length_band": "short", "state_name": "sour...
Invoke the candidate to determine the available resources from the available resources.. Capabilities: list sources. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: none. Side effects: read. Arguments: filters (object optional), scope (string required)
Considering the resource limits and lifecycle, what content should we examine next? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2}, {"action_family": "tool_use", ...
Apply this capability when the next decision depends on the relevant content.. Capabilities: search content, retrieve passages. Accepts: text, pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: indexed sources required. Side effects: read. Arguments: limi...
Given stale governance checks and multiple schema-ambiguous sources, which read-only operation should I run next to locate the exact entity-bound token without writing local state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step...
Apply this capability when the next decision depends on an exact name, code, or token.. Capabilities: exact pattern search. Accepts: text, code. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: textual source required. Side effects: read. Arguments: case sen...
After inspecting full context from eight sources and confirming the acronym issue, what remaining read operation should I perform to select the latest value for the exact identifier? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "latest_value_...
This candidate supports controlled inspection of resource metadata and filters.. Capabilities: search metadata. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: metadata catalog required. Side effects: read. Arguments: filters (object required), sort (string op...
When comparing schema-bound candidates from Excel, should I extract field/type structure or instead read the raw content to resolve the long-document needle, given the transformation goal? Decision: route Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "terminal_readiness", "question_length_band"...
Use this candidate when the task requires fields, types, and structure.. Capabilities: inspect structured schema. Accepts: csv, excel, sqlite. Returns: schema. Evidence role: structure discovery. Prerequisites: source selected. Constraints: structured source required. Side effects: read. Arguments: include types (boole...
To verify the synthesized access rules, which read-only operation should I run to find the matching records across the contradictory evidence sources? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false,...
This operation handles matching records while preserving the current task state.. Capabilities: search structured records. Accepts: csv, excel, sqlite. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: schema known. Constraints: structured source required. Side effects: read. Arguments: l...
With only two steps left, should I prioritize scanning the document for section headings and structural markers to verify the absence of the disputed requirement? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"act...
Invoke the candidate to determine sections and structural landmarks from the available resources.. Capabilities: inspect document structure. Accepts: pdf. Returns: document structure. Evidence role: structure discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: lan...
After that incomplete attempt, what should I do next to find the exact pages or ranges that contain the required data transformation, given the paraphrase obstacle? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "latest_value_selection"} Agent ...
The operation produces a reliable result for the relevant pages or ranges under current constraints.. Capabilities: search document pages. Accepts: pdf. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments: page...
Which resource should I select next to look up the access rules and permissions, given the long-document needle issue and partial source inventory? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "o...
Apply this capability when the next decision depends on the selected resource.. Capabilities: read content. Accepts: text, code. Returns: content. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: local source required. Side effects: read. Arguments: offset (integer optional), resource (s...
What should I do next to enumerate all sources that define the relevant symbols and definitions for consistency checks, using metadata filters? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "source_d...
This candidate supports controlled inspection of symbols and definitions.. Capabilities: inspect code structure. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Arguments: depth (integer required)...
Given the retrieval phase and absence verification, should I inspect a candidate result to check if the exact identifier is missing? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "abse...
Use this candidate when the task requires a candidate result.. Capabilities: inspect evidence. Accepts: evidence candidates. Returns: evidence. Evidence role: evidence inspection. Prerequisites: evidence candidate selected. Constraints: candidate evidence required. Side effects: read. Arguments: claim (string optional)...
With snippets only and a disputed requirement, what should I do to get the surrounding context for a candidate? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "contradiction_resolution"} Agent State: {"phase": "retrieval", "question_length_band"...
This operation handles missing surrounding context while preserving the current task state.. Capabilities: expand context. Accepts: evidence candidates, evidence. Returns: evidence. Evidence role: context expansion. Prerequisites: evidence candidate selected. Constraints: incomplete context required. Side effects: read...
To resolve the contradiction between sources, should I compare the candidate results to see if they agree or conflict? Decision: route Plan: {"active": true, "operation": "contradiction_resolution"} Agent State: {"phase": "retrieval", "question_length_band": "long", "state_name": "retrieval"} Source State: {"available_...
Invoke the candidate to determine agreement or conflict between results from the available resources.. Capabilities: compare evidence. Accepts: evidence. Returns: evidence comparison. Evidence role: evidence synthesis. Prerequisites: evidence inspected. Constraints: multiple evidence required. Side effects: none. Argum...
Given partial requirements and governance phase, how should I assess task coverage to decide the next workflow step? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": "weak", "step": 2...
The operation produces a reliable result for task requirements and coverage under current constraints.. Capabilities: update requirements. Accepts: evidence, agent state. Returns: governance state. Evidence role: governance update. Prerequisites: evidence inspected. Constraints: tracked requirements required. Side effe...
Given the partial progress and unresolved items, should I check if the current evidence is sufficient before proceeding, or is a state update needed first? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": f...
Apply this capability when the next decision depends on whether available information is sufficient.. Capabilities: assess evidence. Accepts: evidence, governance state. Returns: assessment. Evidence role: governance assessment. Prerequisites: requirements updated. Constraints: canonical evidence set required. Side eff...
Since requirements are complete but contradictions remain, what should I do next: inspect the full document for the needle or finalize with a terminal result? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active"...
This candidate supports controlled inspection of the selected result or next action.. Capabilities: finalize selection. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh sufficient assessment required. Side effects...
With only snippets seen and a split context, should I retrieve the full records to join them, or is it better to generate a plan for validation? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use",...
Use this candidate when the task requires the best next information-gathering step.. Capabilities: plan retrieval. Accepts: agent state, text. Returns: plan. Evidence role: none. Prerequisites: objective available. Constraints: none. Side effects: local state write. Arguments: objective (string required), priority (str...
Given stale assessments and near-duplicate sources, how should I verify the absence of a version? Should I inspect the PDF content or compare evidence candidates? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"act...
This operation handles the available resources while preserving the current task state.. Capabilities: list sources. Accepts: metadata. Returns: source list. Evidence role: source discovery. Prerequisites: none. Constraints: none. Side effects: read. Arguments: filters (object optional), scope (string required)