github-evidence-kit

Generate, export, load, and verify forensic evidence from GitHub sources. Use when creating verifiable evidence objects from GitHub API, GH Archive, Wayback Machine, local git repositories, or security vendor reports. Handles evidence storage, querying, and re-verification against original sources.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/oss-forensics/github-evidence-kit/SKILL.md

Quality

Content

63%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with executable examples and method tables, and benefits from a built-in verification step. Its weaknesses are token-efficiency padding, the absence of an explicit verify-and-retry loop, and a monolithic structure that inlines reference material rather than splitting it across the available bundle files.

Suggestions

Trim padding: remove the redundant 'commit = collector.collect_commit("HEAD")' / 'abc123' duplicate assignment and the 'Forensic gold!' exclamatory comments, and shorten the WorkflowRunEvent table row to a one-line availability caveat.

Add an explicit fix-and-retry loop after verification (e.g. 'if not is_valid: inspect errors, re-collect the affected evidence, and re-run store.verify_all() until clean') to reach full workflow-clarity.

Move the per-collector API method tables and/or the GCP credential walkthrough into a references/ file and link to it, and surface the existing scripts/ingest_bq_events.py and scripts/init_investigation.py from the body so the bundle is discoverable.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient (tables plus code), but includes padding that could be trimmed — redundant 'commit = collector.collect_commit(...)' re-assignments, 'Forensic gold!' exclamations, and a verbose WorkflowRunEvent caveat row — so it sits at 'mostly efficient but could be tightened' rather than the lean 4.	3 / 5
Actionability	Code blocks are executable and copy-paste ready with concrete arguments, and method tables give full signatures, but several collectors (GitHubAPICollector methods, WaybackCollector) only have table signatures without usage examples, leaving minor gaps short of a 5.	4 / 5
Workflow Clarity	Quick Start sequences collect → store → save → verify with the verify step acting as a checkpoint, and verification is core to the skill so the destructive/batch cap-at-3 does not apply; however no explicit fix-and-retry recovery loop is shown, capping it just below 5.	4 / 5
Progressive Disclosure	The ~340-line body has good section headers but inlines content that could live in separate files (per-collector API reference, evidence-type catalog, GCP credential setup), and the existing scripts/ bundle files (ingest_bq_events.py, init_investigation.py) are not referenced or surfaced, so it is 'some structure but content that should be separate is inline'.	3 / 5
	Total	14 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description that explicitly covers both capability and trigger conditions with concrete, source-specific language. The only weakness is trigger-term breadth — the terms are accurate but technical, missing natural synonyms and file extensions users might say.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — 'Generate, export, load, and verify forensic evidence' plus 'Handles evidence storage, querying, and re-verification' — giving comprehensive coverage rather than the 1-2 actions of a 3.	5 / 5
Completeness	It explicitly answers both what ('Generate, export, load, and verify forensic evidence... Handles evidence storage, querying, and re-verification') and when ('Use when creating verifiable evidence objects from GitHub API, GH Archive, Wayback Machine, local git repositories, or security vendor reports') with concrete trigger phrases.	5 / 5
Trigger Term Quality	It names concrete sources ('GitHub API, GH Archive, Wayback Machine, local git repositories, security vendor reports') giving good keyword coverage, but leans technical and omits synonyms/file extensions (e.g. 'evidence.json', '.json'), stopping short of a 5.	4 / 5
Distinctiveness Conflict Risk	The forensic-evidence-from-GitHub-sources niche with named source feeds is clearly distinct and unlikely to trigger for unrelated skills, fitting the clear-niche/minimal-conflict anchor.	5 / 5
	Total	19 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: gadievron/raptor
Path: .claude/skills/oss-forensics/github-evidence-kit/SKILL.md
Commit: a5deef3

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.