recall
Fan-out search across all memory sources when context is unclear or vaguely referenced. Triggers on: 'from earlier', 'remember when', 'what we discussed', 'that thing with', 'the conversation about', 'did we ever', 'what happened with', 'you mentioned', 'we talked about', 'earlier today', 'last session', 'the other day', or any vague reference to past context that needs resolution before the agent can act.
88
83%
Does it follow best practices?
Impact
96%
1.41xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Quality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description with excellent trigger term coverage and clear completeness, explicitly listing both what the skill does and when it should be used. Its main weakness is that the 'what' portion is somewhat thin — it describes the search strategy ('fan-out search across all memory sources') but doesn't enumerate specific concrete actions or outputs beyond that. The trigger phrase list is exceptionally thorough and natural.
Suggestions
Add more specific concrete actions/outputs beyond 'fan-out search', e.g., 'retrieves conversation history, cross-references session logs, surfaces relevant notes and decisions, and synthesizes matching context from multiple memory stores.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain ('fan-out search across all memory sources') and a general action ('search' when 'context is unclear or vaguely referenced'), but it doesn't list multiple concrete actions beyond searching. It's more about the trigger conditions than specific capabilities like 'retrieve conversation logs, search notes, cross-reference session history.' | 2 / 3 |
Completeness | Clearly answers both 'what' (fan-out search across all memory sources when context is unclear) and 'when' (explicit trigger phrases listed, plus the general condition 'any vague reference to past context that needs resolution before the agent can act'). The 'Triggers on:' clause serves as an explicit 'Use when' equivalent. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger phrases users would actually say: 'from earlier', 'remember when', 'what we discussed', 'that thing with', 'the conversation about', 'did we ever', 'you mentioned', 'we talked about', 'earlier today', 'last session', 'the other day'. These are highly natural and comprehensive. | 3 / 3 |
Distinctiveness Conflict Risk | This skill occupies a clear niche — vague references to past context requiring multi-source memory search. The extensive list of specific trigger phrases and the qualifier 'when context is unclear or vaguely referenced' make it highly distinguishable from other memory or search skills that might handle specific, well-defined queries. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted, highly actionable skill that provides concrete commands for each memory source and a clear fan-out workflow with mandatory timeout constraints. Its main strength is the specificity of the search pattern and the anti-patterns section. Minor weaknesses include a somewhat lengthy trigger detection list and all content being inline rather than leveraging any bundle structure for the detailed search source reference.
Suggestions
Consider condensing the trigger detection list into a shorter summary pattern (e.g., 'any vague temporal or pronominal reference to past context') rather than enumerating 15+ phrases, since Claude can generalize from fewer examples.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but has some redundancy — the trigger detection list is quite long and could be condensed, and some explanatory text like 'fastest, most likely' and 'Key principle' are mildly verbose. However, it mostly avoids explaining things Claude already knows. | 2 / 3 |
Actionability | Highly actionable with concrete, executable bash commands and tool invocations for each search source. The keyword extraction example ('Those photos from earlier' → keywords: photos, images, media, telegram) is specific and practical. Commands are copy-paste ready with real paths. | 3 / 3 |
Workflow Clarity | Clear 5-step workflow with explicit sequencing, mandatory timeout requirements acting as validation checkpoints, and an explicit error recovery path ('If nothing found — say so honestly'). The anti-patterns section adds guardrails against common failure modes. The fan-out search is clearly ordered by priority. | 3 / 3 |
Progressive Disclosure | Content is well-structured with clear headers and logical sections, but everything is in a single file with no bundle files for reference. The 8 search sources could potentially be split into a reference file, though for a skill of this size (~100 lines) it's borderline acceptable. No external references are provided or needed critically. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- joelhooks/joelclaw
- Commit
03f0a59
- Reviewed
Table of Contents
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.