About

Imported from majiayu000/claude-skill-registry (skills/documents/evidence-logging/SKILL.md). Install upstream with npx skills add majiayu000/claude-skill-registry --skill evidence-logging. Copyright stays with the author.

Evidence Logging

Name: evidence-logging
Author: majiayu000

When to Use

During any review or analysis workflow to capture reproducible evidence.
When findings must be traceable to specific commands, outputs, or sources.
Before finalizing recommendations that stakeholders will act upon.

Activation Patterns

Trigger Keywords: evidence, proof, trace, audit, reproducible, citation, source, verify Contextual Cues:

"show your work" or "provide evidence"
"how can I verify this" or "reproduce these findings"
"cite your sources" or "where did this come from"
"create an audit trail"
"document the steps taken"

Auto-Load When: Any analysis requires traceability or when findings need reproducible verification.

Required TodoWrite Items

evidence-logging:log-initialized
evidence-logging:commands-captured
evidence-logging:citations-recorded
evidence-logging:artifacts-indexed

Mark each item complete as you finish the corresponding step.

Step 1: Initialize Log (`evidence-logging:log-initialized`)

Create evidence structure with timestamp and context:
- Session ID or review identifier.
- Repository, branch, and commit hash.
- Analyst identity and review scope.
Establish naming convention for evidence references (e.g., [E1], [E2]).

Step 2: Capture Commands (`evidence-logging:commands-captured`)

Log every command that produces evidence:

[E1] Command: git diff --stat HEAD~5..HEAD
     Output: 15 files changed, 234 insertions(+), 89 deletions(-)
     Timestamp: 2024-01-15T10:30:00Z

Include full command with arguments (no aliases).
Capture relevant output snippets, not entire dumps.
Note working directory and environment if relevant.

Step 3: Record Citations (`evidence-logging:citations-recorded`)

Log external sources consulted:

[C1] Source: https://doc.rust-lang.org/nomicon/
     Section: "Working with Unsafe"
     Relevance: Validates unsafe block justification

Include web searches performed and key results.
Reference documentation versions (API docs, RFCs, specs).
Note any AI-assisted analysis with model/prompt context.

Step 4: Index Artifacts (`evidence-logging:artifacts-indexed`)

Catalog generated artifacts:
- Screenshots, diagrams, or visualizations.
- Exported reports or coverage summaries.
- Saved query results or API responses.
Provide file paths or inline content for small artifacts.
Note artifact retention policy (ephemeral vs. archived).

Evidence Reference Format

Use consistent format in findings:

Finding: Memory leak in connection pool [E3, C2]
- Evidence [E3]: valgrind output showing 4KB unreleased
- Citation [C2]: PostgreSQL docs on connection lifecycle

Exit Criteria

Todos completed with structured evidence log.
All findings traceable to specific evidence references.
Downstream reports can cite evidence without re-running commands.

Use it

Copy one of these into your project. Installing also returns the manifest and these snippets.

yaml

targets:
  - https://api.opensmartroute.ai/api/v1/registry/majiayu000-claude-skill-registry-evidence-logging/manifest   # or paste the manifest below

Manifest

An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.

majiayu000-claude-skill-registry-evidence-logging.ocm.jsonjson

{
  "ocm": "1",
  "id": "majiayu000-claude-skill-registry-evidence-logging",
  "kind": "skill",
  "name": "evidence-logging",
  "description": "Workflow for capturing evidence and citations to create reproducible analyses and audit trails. Triggers: evidence capture, citations, reproducible analysis, audit trail, documentation, evidence logging, findings documentation Use when: conducting any review that needs evidence trails, creating audit documentation, ensuring reproducibility of analyses DO NOT use when: quick informal checks without documentation needs. DO NOT use when: structured output is the focus - use structured-output. Use this skill as foundation for all evidence-based review workflows.",
  "publisher": "majiayu000",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "general"
    ],
    "tags": [
      "skill-md",
      "evidence",
      "citations",
      "reproducible",
      "audit-trail",
      "documentation",
      "github"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "Workflow for capturing evidence and citations to create reproducible analyses and audit trails. Triggers: evidence capture, citations, reproducible analysis, audit trail, documentation, evidence logging, findings documentation Use when: conducting any review that needs evidence trails, creating audit documentation, ensuring reproducibility of analyses DO NOT use when: quick informal checks without documentation needs. DO NOT use when: structured output is the focus - use structured-output. Use this skill as foundation for all evidence-based review workflows."
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "github",
      "repository": "https://github.com/majiayu000/claude-skill-registry",
      "path": "skills/documents/evidence-logging/SKILL.md",
      "ref": "a16e8e7bdddc9a7406a56f51e520cef7a8646208",
      "url": "https://github.com/majiayu000/claude-skill-registry/blob/a16e8e7bdddc9a7406a56f51e520cef7a8646208/skills/documents/evidence-logging/SKILL.md",
      "key": "majiayu000/claude-skill-registry/skills/documents/evidence-logging/SKILL.md"
    }
  },
  "instructions": "# Evidence Logging\n\n## When to Use\n- During any review or analysis workflow to capture reproducible evidence.\n- When findings must be traceable to specific commands, outputs, or sources.\n- Before finalizing recommendations that stakeholders will act upon.\n\n## Activation Patterns\n**Trigger Keywords**: evidence, proof, trace, audit, reproducible, citation, source, verify\n**Contextual Cues**:\n- \"show your work\" or \"provide evidence\"\n- \"how can I verify this\" or \"reproduce these findings\"\n- \"cite your sources\" or \"where did this come from\"\n- \"create an audit trail\"\n- \"document the steps taken\"\n\n**",
  "cost": {
    "context_tokens": 726
  }
}

Fetch it by URL: GET /api/v1/registry/majiayu000-claude-skill-registry-evidence-logging/manifest?version=1.0.0

Reviews

Star ratings from people who tried it. One review per account; edit yours any time.

No reviews yet. Install it, try it, and be the first to rate it.