Skip to main content
Glama

misakanet_search

Read-onlyIdempotent

Search a public failure-lesson index by error text, keyword, or topic to discover relevant fix lessons when you do not already know a lesson ID.

Instructions

Search MisakaNet's public failure-lesson index by error text, keyword, or topic. Use when you need to discover relevant lessons and do not already know a lesson ID.

Choosing among the three retrieval tools: no error yet and a task about to start → memory_context; an actual error or keyword in hand → search; something risky about to run and no error yet → preflight. Not for: loading context before a task (that is memory_context) or a risk verdict (that is preflight). Input semantics: query is required; domain optionally filters by lesson domain; top limits ranked results and defaults to 5. kind filters by result type: 'lessons' (lesson files only), 'evidence' (results with evidence_refs or verification), 'related' (cross-referenced/tag-overlap), 'all' (default). kind is auto-detected from query intent when omitted (e.g. 'lesson about X' → lessons, 'evidence for X' → evidence). Set explain=true to return matched terms, TF-IDF, entity matches, vector similarity, and hybrid score components. detail controls progressive disclosure: compact (default, ~80 tok/lesson) for broad scans, summary (~200 tok) with domain/tags/fix, full for complete lesson markdown. Output schema: JSON with results[] and source; each result is a ranked lesson summary. Error cases: missing query, unavailable search index, or no matches (empty results). Side effects: none. Auth: none. Rate limits: local stdio process only; callers should keep result counts small. Do not use for private log collection; search only with redacted snippets. Use misakanet_get_lesson for full content.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "query": "<query>"
    },
    "name": "misakanet_search"
  }
}

Input Schema

NameRequiredDescriptionDefault
topNoMaximum ranked results to return. Defaults to 5; keep small for MCP context and latency.
kindNoFilter results by kind: 'lessons' returns only lesson files, 'evidence' returns results with evidence_refs or high evidence_level, 'related' returns cross-referenced/tag-overlap results. Default 'all' returns everything. Auto-detected from query intent when omitted (e.g. 'lesson about X' → lessons, 'evidence for X' → evidence).
queryYesRequired redacted error message, keyword, or topic (for example: 'pip install timeout' or 'DCO sign-off failed').
detailNoProgressive disclosure: compact (default, ~80 tok/lesson) shows id/title/problem/freshness; summary (~200 tok) adds domain/tags/fix; full returns complete lesson markdown. Use compact for broad scans, full only after narrowing results.
domainNoOptional domain filter such as devops, python, network, feishu, rag, fanuc, or mcp.
explainNoInclude score evidence for each result; vector similarity is null when the optional backend is unavailable.
bm25_weightNoOverride BM25 keyword weight (0-1). Higher values favor exact keyword matches. Default: 0.65. All weights must sum to 1.0.
include_staleNoInclude stale and superseded lessons in results. Default false — these are filtered out to avoid误导 agents with outdated information.
baseline_weightNoOverride baseline score weight (0-1). Higher values favor proven/popular lessons. Default: 0.15.
metadata_weightNoOverride metadata bonus weight (0-1). Higher values favor lessons with matching domain/tags. Default: 0.20.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv2.28.0
    • addedInput schema / properties / include_stale
      Added value: +{
      +  "description": "Include stale and superseded lessons in results. Default false — these are filtered out to avoid误导 agents with outdated information.",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / kind
      Added value: +{
      +  "description": "Filter results by kind: 'lessons' returns only lesson files, 'evidence' returns results with evidence_refs or high evidence_level, 'related' returns cross-referenced/tag-overlap results. Default 'all' returns everything. Auto-detected from query intent when omitted (e.g. 'lesson about X' → lessons, 'evidence for X' → evidence).",
      +  "enum": [
      +    "all",
      +    "lessons",
      +    "evidence",
      +    "related"
      +  ],
      +  "type": "string"
      +}
  2. Changed3 schema fields changedv2.23.0
    • addedInput schema / properties / baseline_weight
      Added value: +{
      +  "description": "Override baseline score weight (0-1). Higher values favor proven/popular lessons. Default: 0.15.",
      +  "type": "number"
      +}
    • addedInput schema / properties / bm25_weight
      Added value: +{
      +  "description": "Override BM25 keyword weight (0-1). Higher values favor exact keyword matches. Default: 0.65. All weights must sum to 1.0.",
      +  "type": "number"
      +}
    • addedInput schema / properties / metadata_weight
      Added value: +{
      +  "description": "Override metadata bonus weight (0-1). Higher values favor lessons with matching domain/tags. Default: 0.20.",
      +  "type": "number"
      +}
  3. Changed1 schema field changedv2.21.0
    • addedInput schema / properties / detail
      Added value: +{
      +  "description": "Progressive disclosure: compact (default, ~80 tok/lesson) shows id/title/problem/freshness; summary (~200 tok) adds domain/tags/fix; full returns complete lesson markdown. Use compact for broad scans, full only after narrowing results.",
      +  "enum": [
      +    "compact",
      +    "summary",
      +    "full"
      +  ],
      +  "type": "string"
      +}
  4. Changed1 schema field changedv2.17.1
    • addedInput schema / properties / explain
      Added value: +{
      +  "description": "Include score evidence for each result; vector similarity is null when the optional backend is unavailable.",
      +  "type": "boolean"
      +}
  5. Changed3 schema fields changedv2.14.0
    • changedInput schema / properties / domain / description
      Previous value: -"Optional domain filter (devops, python, network, feishu, rag, fanuc, etc.)"New value: +"Optional domain filter such as devops, python, network, feishu, rag, fanuc, or mcp."
    • changedInput schema / properties / query / description
      Previous value: -"Search query — error message, keyword, or topic (e.g. 'pip install timeout', 'DCO sign-off failed')"New value: +"Required redacted error message, keyword, or topic (for example: 'pip install timeout' or 'DCO sign-off failed')."
    • changedInput schema / properties / top / description
      Previous value: -"Max results to return (default 5)"New value: +"Maximum ranked results to return. Defaults to 5; keep small for MCP context and latency."
  6. Addedv2.12.4
  7. Removedv2.12.2
  8. First observedv2.12.3

TDQS

A4.7/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the readOnly/idempotent/OpenWorld annotations by stating side effects (none), auth (none), the local stdio execution model with a caller-side rate/count caution, enumerated error cases, and the output envelope (results[] and source). It also discloses a real behavioral rule — stale and superseded lessons are filtered by default and require include_stale — and a privacy guard ('do not use for private log collection').

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Long but deliberately structured: routing rule first, then exclusions, then input semantics, then output, errors, side effects, and rate limits. Each block is scannable, with nothing decorative; the only mild waste is the near-duplication of enum meanings already in the schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 10-parameter hybrid retrieval tool with no output schema, the definition covers invocation guidance, parameter intent, return shape, failure modes, and trust boundaries. The only gap — score-weight interactions — is documented in the schema itself, which is complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so most description text about query, domain, top, kind, and detail restates the schema nearly verbatim, adding little meaning. The description does not mention the three scoring-weight parameters (bm25_weight, baseline_weight, metadata_weight) or the sum-to-1.0 constraint at all, leaving the schema to carry them. Baseline 3 is appropriate since the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('search MisakaNet's public failure-lesson index') plus the three retrieval modes by which to route. It explicitly distinguishes itself from siblings misakanet_memory_context, misakanet_preflight, and misakanet_get_lesson, so an agent can choose without opening schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives an explicit decision rule ('no error yet and a task about to start → memory_context; an actual error or keyword in hand → search; something risky about to run → preflight'), a 'Not for' section naming the alternatives, and a hand-off to misakanet_get_lesson for full content. When-to-use and when-not-to-use are both covered.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.