anchor
stringlengths
723
2.15k
positive
stringlengths
153
8.2k
Given only two steps left, which operation should I prioritize to cover all four unresolved 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}, {"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...
The last attempt was incomplete and the assessment is stale; what should I do next to recover and verify sufficiency? Decision: route Plan: {"active": false, "operation": "lookup_semantic"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fresh_sufficient"} Source State: {"ava...
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...
I have one step remaining; which available resource should I consult to finalize the schema-related decision? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "enumerate_sources"} Agent State: {"phase": "retrieval", "question_length_band": "long",...
This candidate supports controlled inspection of the selected result or next action.. Capabilities: finalize selection, emit terminal result. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh sufficient assessment ...
What is the next action to verify absence of evidence for the schema field? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "absence_verification"} Agent State: {"phase": "retrieval", "question_length_band": "short", "state_name": "retrieval"} So...
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...
Which available resource should I use next to satisfy the telemetry and trace requirements? Decision: route Plan: {"active": true, "operation": "compatibility"} Agent State: {"phase": "planning", "question_length_band": "short", "state_name": "initial"} Source State: {"available_modalities": ["excel", "metadata"], "evi...
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)
Can you point me to the part of the document that matches the trace requirements? 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": "governance", "questi...
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...
What is the exact name or token for the field definition that fits the schema? Decision: route Plan: {"active": false, "operation": "lookup_semantic"} Agent State: {"phase": "inspection", "question_length_band": "short", "state_name": "partial_evidence"} Source State: {"available_modalities": ["sqlite", "metadata"], "e...
Invoke the candidate to determine an exact name, code, or token from the available resources.. 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...
Considering the disputed requirements, which resource metadata or filter should be applied to resolve the contradiction? 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...
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...
Given the disputed requirements and mixed sources, what read-only operation should I run next to compare field types and structure without writing any state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "synt...
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...
Since I've already checked the CSV and metadata, what remains to join the matching records for version compatibility, given the acronym obstacle? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "join"} Agent State: {"phase": "terminal_readiness"...
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 ...
Instead of just extracting content, which structural operation should I perform next to locate the sections and landmarks in the mixed sources, given the schema ambiguity? Decision: route Plan: {"active": true, "operation": "compatibility"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name"...
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...
How can I verify the consistency across sources by enumerating the relevant pages or ranges in the PDFs, using a read-only operation? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "enu...
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 (...
We have one step left. Which resource should I use to compare the complementary sources and check consistency? Decision: route Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "synthesis", "question_length_band": "long", "state_name": "contradiction"} Source State: {"available_modalities": ["text"...
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...
The previous attempt was incomplete due to an acronym. I need to resolve field definitions and compatibility. What should I do next? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phase": "inspection", "...
Apply this capability when the next decision depends on symbols and definitions.. Capabilities: inspect code structure, resolve symbols. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Arguments: ...
I need to perform a data transformation. Which candidate result should I use, given the version noise and stale assessment? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "compare"} Age...
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...
What is the next workflow step? I need the surrounding context to resolve a contradiction. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "contradiction_resolution"} Agent State: {"phase": "planning", "question_length_band": "short", "state_name...
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...
Based on the inspected multi-source evidence, what operation should determine if the join results agree or conflict? 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...
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...
What should be done next to verify that the workflow step meets 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_family": "tool_use...
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...
Given the stale assessment and unknown sources, what operation would determine if the available information is sufficient for consistency checks? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "ope...
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...
Given the contradictory evidence and partial context, what next action should verify the absence of the expected event delivery? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "absence_verification"} Agent State: {"phase": "terminal_readiness",...
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 ...
Given the known inventory and broad query, which read-only step should I take next to enumerate sources while avoiding any state writes? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "...
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),...
I've only checked a couple of things and haven't inspected any evidence yet. What resources are still available to compare against, given the mixed formats and unknown inventory? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "compare"} Agent St...
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)
Should I retrieve the exact identifier's content now, or would it be premature to inspect the stale assessment? I need the relevant content from the PDF. Decision: route Plan: {"active": true, "operation": "lookup_semantic"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name": "initial"} Sou...
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...
To verify the exact token against contradictory evidence, which read-only operation should I use to get a source list without side effects? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "lookup_semantic"} Agent State: {"phase": "governance", "...
Invoke the candidate to determine an exact name, code, or token from the available resources.. Capabilities: exact pattern search, identifier search. Accepts: text, code. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: textual source required. Side effects:...
Given limited steps, what should I prioritize next to enumerate sources and resolve disputed requirements, considering resource metadata and filters? Decision: route Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fres...
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...
After a failed attempt, how do I recover to discover sources for field definitions and compatibility, given schema ambiguity and no prior evidence? 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 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...
Which catalog operation should I perform next to find matching records for the technical instruction, focusing on latest value selection? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "resul...
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...
What should I do next to locate sections and structural landmarks for failure recovery, given contradictory evidence and limited steps? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phase": "terminal_re...
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 ...
Which pages or ranges in the source documents are relevant for validating the data quality constraints? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "latest_value_selection"} Agent St...
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...
Can you point me to the selected resource that addresses the 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 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...
What symbols and definitions in the code are needed to perform the required data 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}, {"action_family":...
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...
Given the partial progress and known source inventory, which candidate result should be used for the next workflow step? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "enumerate_sources"} Agent State: {"phase": "terminal_readiness", "question_...
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...
We have only snippets from one PDF and need to enumerate all records, but identifiers vary. What should we do next to get the missing surrounding context without writing any 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 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...
We've inspected multiple sources but haven't seen the expected absence. Given the disputed requirement and remaining steps, what should we check next to confirm agreement or conflict between the results? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", ...
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...
We have two evidence snippets that seem contradictory. We need to resolve this, but we shouldn't just pick one. What operation should we do next to cover task requirements, instead of a simple content retrieval? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "too...
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...
We've exhausted the sources and have one piece of evidence. How can we verify whether the available information is sufficient for failure recovery, without making any state changes? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "ste...
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...
With only two steps left, which next action should I take to resolve the acronym issue and find the exact entity-bound match? Decision: route Plan: {"active": false, "operation": "lookup_exact"} Agent State: {"phase": "source_discovery", "question_length_band": "long", "state_name": "source_discovery"} Source State: {"...
Invoke the candidate to determine the selected result or next action from the available resources.. Capabilities: finalize selection. Accepts: assessment, governance state. Returns: terminal result. Evidence role: terminal decision. Prerequisites: requirements complete. Constraints: fresh sufficient assessment required...
My first attempt failed due to paraphrasing; what should I do next to resolve the contradiction and get the schema-bound info? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "contradiction_resolution"} Agent State: {"phase": "retrieval", "questi...
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...
What resources are available to perform the consistency check across sources, given the schema ambiguity? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "lookup_exact"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long"...
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)
What should I do next to find the relevant technical instruction in the long document? 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...
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...
Find the exact identifier for the required transformation from the inspected sources. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "lookup_semantic"} Agent State: {"phase": "planning"...
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...
Which technical instruction should be followed to compare the schema-bound resources? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "governance", "question_length_band": "short", "state_name": "insufficient"} ...
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...
What fields and types are needed to proceed with the workflow? 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", "result": "weak", "st...
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:...
How should the failure be handled by joining matching records from the inspected evidence? 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...
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...
Given the known PDF sources and partial progress, what read-only operation should identify sections and structural landmarks to support version compatibility comparison? Decision: route Plan: {"active": false, "operation": "compare"} Agent State: {"phase": "source_discovery", "question_length_band": "long", "state_name...
This candidate supports controlled inspection of sections and structural landmarks.. Capabilities: inspect document structure, discover sections. Accepts: pdf. Returns: document structure. Evidence role: structure discovery. Prerequisites: source selected. Constraints: pdf source required. Side effects: read. Arguments...
Since we've checked the full context and have one unresolved requirement, what read-only step should now find the relevant pages or ranges to determine the next workflow step? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}...
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...
For the exact identifier in the SQLite source, should we perform a read-only lookup of the selected resource's telemetry and trace requirements, rather than writing state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Pl...
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...
To verify the disputed requirement, which read-only operation should enumerate symbols and definitions from the mixed sources to confirm resource limits and lifecycle? Decision: route Plan: {"active": false, "operation": "enumerate_records"} Agent State: {"phase": "inspection", "question_length_band": "long", "state_na...
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...
Given 4 remaining steps, which candidate should I inspect next to resolve the disputed requirement? Decision: route Plan: {"active": false, "operation": "lookup_exact"} Agent State: {"phase": "inspection", "question_length_band": "short", "state_name": "partial_evidence"} Source State: {"available_modalities": ["excel"...
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...
After the partial join, what evidence should I retrieve next to fill in the missing context? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "join"} Agent State: {"phase": "synthesis", "question_length_band": "short", "state_name": "contradictio...
Apply this capability when the next decision depends on missing surrounding context.. Capabilities: expand context, resolve split context. Accepts: evidence candidates, evidence. Returns: evidence. Evidence role: context expansion. Prerequisites: evidence candidate selected. Constraints: incomplete context required. Si...
Which available resource should I check to verify the absence and 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_use", "result": "weak", "step": 2}, {"action_family": "tool_us...
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...
What should I do next to enumerate sources for the schema ambiguity? 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", "result": "weak...
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...
We have two near-duplicate documents and need the latest value. Have we already gathered enough evidence to pick it, or should we inspect further? Decision: route Plan: {"active": true, "operation": "latest_value_selection"} Agent State: {"phase": "retrieval", "question_length_band": "long", "state_name": "retrieval"} ...
Use this candidate when the task requires 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: local st...
We've verified absence across all sources. What should we do next: report the result or update our records? Decision: route Plan: {"active": false, "operation": "absence_verification"} Agent State: {"phase": "governance", "question_length_band": "long", "state_name": "insufficient"} Source State: {"available_modalities...
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...
To resolve the acronym and find the exact field definition, what's the next step we should take to gather information? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "lookup_exact"} Age...
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...
Given the stale assessment and the need to enumerate records from Excel, what resources are available to proceed? 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"} Ag...
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)
We need to decide the next step for event delivery retry behavior. We have only snippets from the PDF and haven't inventoried sources. What should we do next to find the relevant content, considering safety and side effects? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"a...
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...
We've inspected multiple sources and found contradictions. We need to resolve them by finding an exact name or token in the code. What should we do next? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "t...
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 ...
We need to check compatibility of the technical instruction. We have full context and fresh assessment. Should we filter by metadata or instead do a full content comparison? What's the precise next move? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", ...
Use this candidate when the task requires 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 optional)
We need to verify absence of a field in the CSV structure. We have snippets and partial progress. How should we confirm the schema doesn't include that field, considering we've seen near-duplicates? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "resu...
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:...
Given only one step left and partial requirements, which candidate should I prioritize to match records and locate the technical instruction? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "r...
The operation produces a reliable result for matching records under current constraints.. Capabilities: search structured records, filter records. Accepts: csv, excel, sqlite. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: schema known. Constraints: structured source required. Side eff...
After an incomplete attempt with only snippets, how do I recover to find sections and structural landmarks for cross-source consistency? Decision: route Plan: {"active": false, "operation": "join"} Agent State: {"phase": "synthesis", "question_length_band": "long", "state_name": "contradiction"} Source State: {"availab...
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...
For event handling, which resource or catalog operation should I use to locate the relevant pages or ranges, given stale assessments and complete requirements? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_famil...
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...
What is the next action to verify field definitions and compatibility for the selected resource, given complementary sources and a partial inventory? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_...
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...
Check the field definitions and compatibility for the schema-bound identifiers. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "absence_verification"} Agent State: {"phase": "retrieval", "question_length_band": "short", "state_name": "retrieval"...
This operation handles symbols and definitions while preserving the current task state.. Capabilities: inspect code structure, resolve symbols. Accepts: code. Returns: content, code structure. Evidence role: evidence inspection. Prerequisites: source selected. Constraints: code source required. Side effects: read. Argu...
List all candidate results that match the entity-bound query, considering the resource limits. Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "enumerate_records"} Agent State: {"phase": "retrieval", "question_length_band": "short", "state_name":...
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...
Find the missing surrounding context needed to join the code-based evidence. Decision: route Plan: {"active": true, "operation": "join"} Agent State: {"phase": "planning", "question_length_band": "short", "state_name": "initial"} Source State: {"available_modalities": ["code", "metadata"], "evidence_topology_observed":...
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...
Compare the evidence results to see if they agree or conflict, given the schema-bound lookup. Decision: route Plan: {"active": false, "operation": "lookup_exact"} Agent State: {"phase": "inspection", "question_length_band": "short", "state_name": "partial_evidence"} Source State: {"available_modalities": ["csv", "metad...
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...
Given the disputed requirement and partial context, what read-only step should I take next to ensure coverage without writing local state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation":...
Apply this capability when the next decision depends on 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 stat...
We've already checked the schema and metadata, but the assessment is stale and two requirements remain. What should I check next to see if the available info suffices? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"acti...
Use this candidate when the task requires whether available information is sufficient.. Capabilities: assess evidence, validate sufficiency. Accepts: evidence, governance state. Returns: assessment. Evidence role: governance assessment. Prerequisites: requirements updated. Constraints: canonical evidence set required. ...
Instead of just picking the latest version, what should I do next to identify the correct source, given the inventory is unknown and the query is broad? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": fals...
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...
How can I verify the join result is correct by gathering more evidence, given the schema ambiguity and disputed 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", ...
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...
With only 4 steps left, which operation should I prioritize to resolve the contradictory evidence and move toward the resource limits and lifecycle requirements? Decision: route Plan: {"active": true, "operation": "latest_value_selection"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name":...
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)
The last attempt was incomplete and ambiguous; what should I do next to recover and enumerate the records correctly despite the identifier variation? Decision: route Plan: {"active": true, "operation": "enumerate_records"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name": "initial"} Sourc...
The operation produces a reliable result for the relevant content under current constraints.. 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 (i...
I need to compare resources but the catalog is only partially known; what operation should I use to find the exact code or token among the near-neighbor candidates? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "compare"} Agent State: {"phase"...
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...
What should I do next to enumerate the event records with the right metadata filters, considering the version noise? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "enumerate_records"} Agent State: {"phase": "terminal_readiness", "question_leng...
Use this candidate when the task requires resource metadata and filters.. 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 (object optional), sort (...
What should be the next step to enumerate the fields, types, and structure from the available sources? 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_fa...
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...
What transformation is needed to find matching records in the Excel data? 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: {"phase": "source_discover...
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...
What action should we take to locate the sections and structural landmarks for telemetry and trace requirements? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "compatibility"} Agent State: {"phase": "governance", "question_length_band": "long"...
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...
What should be the next step to look up the exact field definitions and compatibility, considering the schema ambiguity, from the PDF sources? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "lookup_exact"} Agent State: {"phase": "governance", "...
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...
Given the acronym ambiguity, which read-only candidate best matches the entity-bound instruction for document analysis? 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"...
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...
We've seen conflicting evidence across sources; what remains to check for version compatibility, considering local state writes are allowed? Decision: route Plan: {"active": true, "operation": "contradiction_resolution"} Agent State: {"phase": "planning", "question_length_band": "long", "state_name": "initial"} Source ...
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)...
Instead of finalizing now, which candidate result should we inspect for resource limits, given the split context? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "lookup_exact"} Agent State: {"phase": "terminal_readiness", "question_length_band"...
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)...
How can we verify the required transformation by comparing evidence candidates that differ only in schema? 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...
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...
With no steps left, should I prioritize confirming agreement between the latest event delivery result and the retry behavior evidence? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": false, "operation": "l...
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...
After the incomplete attempt, what should I do next to cover all validation and quality constraints? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "enumerate_records"} Agent State: {"p...
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...
Before continuing, should I check whether the available sources are sufficient for resolving the version compatibility contradiction? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_use", "result": ...
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...
Given the complete requirements, what should I do next to enumerate the access rule records? 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": "to...
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...
We've checked the basics but haven't confirmed a missing value. What should we inspect next to verify that absence? 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"...
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...
What sources are available to compare for consistency across the PDFs? 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": "synthesis", "question_length_ba...
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)
We need to ensure event delivery and retry behavior are correct. What content should we look at next to verify that? Decision: route Plan: {"active": false, "operation": "absence_verification"} Agent State: {"phase": "terminal_readiness", "question_length_band": "long", "state_name": "fresh_sufficient"} Source State: {...
Invoke the candidate to determine the relevant content from the available resources.. 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...
We have a contradiction about resource limits. What exact name or code should we find to resolve it, given we need to read only? Decision: route Plan: {"active": false, "operation": "contradiction_resolution"} Agent State: {"phase": "synthesis", "question_length_band": "long", "state_name": "contradiction"} Source Stat...
Apply this capability when the next decision depends on an exact name, code, or token.. Capabilities: exact pattern search, identifier search. Accepts: text, code. Returns: evidence candidates. Evidence role: candidate discovery. Prerequisites: query available. Constraints: textual source required. Side effects: read. ...
Given the stale assessment and inspected multi-source evidence, what read-only operation should update resource metadata and filters for the next step? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phas...
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 have partially validated some data and know the source inventory; what remains to check the fields, types, and structure for the next validation step? Decision: route Plan: {"active": false, "operation": "latest_value_selection"} Agent State: {"phase": "governance", "question_length_band": "long", "state_name": "ins...
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...
Although we have full context and several candidates, what structural match should be done next to find matching records, rather than a broad search? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}, {"action_family": "tool_...
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...
With fresh assessment and partial context, how should we verify the result by inspecting sections and structural landmarks in the evidence? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation"...
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...
Given our remaining budget, which step should we prioritize to locate the exact pages or ranges for the data quality validation constraints? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": false, "operation": "lookup_exact"} Agent State: {"phase": "synthesis", "que...
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 (...
Our last attempt was incomplete; how can we recover by enumerating sources to identify the selected resource for failure handling? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}] Plan: {"active": true, "operation": "enumerate_sources"} Agent State: {"phase": "retrieval", "question_...
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...
What is the next workflow step to locate symbols and definitions from the available resources without naming any specific tool? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "lookup_ex...
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 we do next to evaluate a candidate result for the next workflow step, given our current state? Decision: route History: [{"action_family": "tool_use", "result": "weak", "step": 0}, {"action_family": "tool_use", "result": "weak", "step": 1}] Plan: {"active": true, "operation": "compare"} Agent State: {"phase...
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...