auto-capture-claude-code

Claude Code adapter for the auto-capture skill. Extends auto-capture with automatic session-end hooks that capture transcripts to Open Brain without manual intervention. Use this when you want every meaningful Claude Code session to be preserved automatically — not just the ones where you remember to say "wrap up".

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/auto-capture-claude-code/SKILL.md

Quality

Content

77%

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

The content is actionable and workflow-clear with copy-paste-ready hook config and explicit verification steps, but it restates the base-vs-adapter distinction redundantly and references bundle files that are not actually shipped.

Suggestions

Collapse the repeated base-vs-adapter complementarity explanation into a single section to remove the redundancy across 'Relationship to Upstream Skill', 'What This Adapter Adds', and 'Notes'.

Ship session-end-capture.mjs (and config/sensitivity-patterns.json) in a scripts/ or references/ bundle so the signaled one-level-deep references actually resolve, or inline a minimal runnable version.

Tighten 'Notes' to only the non-blocking and tool-agnostic points, since complementarity is already covered earlier.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and well-sectioned, but the base-vs-adapter complementarity point is restated three times (Relationship to Upstream Skill, What This Adapter Adds, and Notes), adding redundancy a tighter pass would remove.	2 / 3
Actionability	It provides a complete, copy-paste-ready hook registration JSON, concrete env-var setup, numbered install steps, and named tunables (RETRY_MAX_ATTEMPTS, HARD_TIMEOUT_MS), giving fully executable guidance.	3 / 3
Workflow Clarity	The How It Works sequence, numbered installation checklist with an explicit verification step (step 5: verify a capture appears), and a described retry-queue feedback loop give a clear sequence with explicit checkpoints.	3 / 3
Progressive Disclosure	Sections are well-organized, but referenced files (session-end-capture.mjs, config/sensitivity-patterns.json, data/capture-retry-queue/, logs/) are not present in any references/, scripts/, or assets/ bundle, so the signaled one-level-deep references are not actually delivered.	2 / 3
	Total	10 / 12 Passed

Description

75%

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

The description is clear and complete with an explicit Use-when trigger and a distinct niche, but its trigger phrasing is desire-oriented rather than mirroring natural user utterances, and the second-person voice incurs a specificity penalty.

Suggestions

Rewrite the trigger in third person and mirror natural user phrasing, e.g. 'Use when a Claude Code session ends without an explicit wrap-up trigger and you want its transcript preserved.'

Surface the concrete trigger verbs users actually say ('wrap up', 'park this', session end) earlier in the description to strengthen trigger-term coverage.

Convert 'Use this when you want...' to third person ('Use when Claude Code sessions end without...') to avoid the second-person specificity penalty.

Dimension	Reasoning	Score
Specificity	The description names concrete actions ('Extends auto-capture with automatic session-end hooks that capture transcripts to Open Brain without manual intervention'), which would score 3, but the second-person voice ('Use this when you want') triggers the mandatory -1 specificity penalty per the rubric guidelines.	2 / 3
Completeness	It explicitly answers both what ('Extends auto-capture with automatic session-end hooks that capture transcripts to Open Brain') and when ('Use this when you want every meaningful Claude Code session to be preserved automatically'), with an explicit Use-when trigger clause.	3 / 3
Trigger Term Quality	It includes some relevant natural terms ('Claude Code session', 'wrap up') but the trigger phrasing is desire-based ('when you want every meaningful... session to be preserved automatically') rather than keywords a user would naturally utter, missing common variations.	2 / 3
Distinctiveness Conflict Risk	It carves a clear niche (Claude Code ambient session-end capture) tied to an explicit upstream skill, making it unlikely to trigger for unrelated skills despite intended overlap with the base auto-capture skill.	3 / 3
	Total	10 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 suspicious	Warning

	Total	14 / 16 Passed

Repository: NateBJones-Projects/OB1
Path: skills/auto-capture-claude-code/SKILL.md
Commit: 6779106

Reviewed: about 17 hours 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.