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.
90
87%
Does it follow best practices?
Impact
96%
1.41xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Quality
Discovery
82%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 description excels at trigger term coverage with an extensive list of natural phrases users would say when referencing past context. The 'when' guidance is exceptionally thorough. However, the 'what' portion could be more specific about the concrete actions performed beyond 'fan-out search,' and the description could better distinguish itself from other potential memory-related skills.
Suggestions
Add more specific concrete actions beyond 'fan-out search' — e.g., 'Searches conversation history, stored notes, and session logs to resolve vague references, then summarizes matching context.'
Clarify distinctiveness by specifying what differentiates this from other memory skills — e.g., 'Use this skill specifically for ambiguous/vague references rather than direct memory lookups by topic or date.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain ('fan-out search across all memory sources') and one core action (searching when context is unclear), but doesn't list multiple concrete actions like retrieving, summarizing, or correlating past conversations. 'Fan-out search' is somewhat technical and the actual capabilities beyond searching are not detailed. | 2 / 3 |
Completeness | Clearly answers both 'what' (fan-out search across all memory sources when context is unclear or vaguely referenced) 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 trigger guidance is thorough and explicit. | 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', 'we talked about', 'last session', 'the other day', etc. These are highly natural and comprehensive variations of how users reference past context. | 3 / 3 |
Distinctiveness Conflict Risk | While the trigger phrases are specific, 'search across all memory sources' could overlap with other memory-related skills (e.g., a specific memory retrieval skill or a conversation history skill). The description doesn't clearly distinguish itself from other memory/recall skills that might exist in a larger skill set. | 2 / 3 |
Total | 10 / 12 Passed |
Implementation
92%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, actionable skill that provides concrete commands for every memory source, a clear workflow with failure handling, and important safety constraints (timeouts, anti-patterns). The trigger detection list is comprehensive and the fan-out pattern is well-structured. The only weakness is that all 8 search sources are inline rather than referenced from a separate file, making the skill slightly longer than ideal for quick scanning.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is lean and efficient. Every section serves a purpose — trigger detection, search sources, workflow, timeouts, anti-patterns. There's no explanation of what grep does or how memory works conceptually. The trigger list is necessarily verbose but earns its place as a reference lookup. | 3 / 3 |
Actionability | Provides fully executable bash commands for each memory source, specific file paths, concrete tool invocations (session_context, redis-cli, slog), and copy-paste ready examples with good/bad timeout patterns. The keyword extraction example ('Those photos from earlier' → keywords) is a concrete worked example. | 3 / 3 |
Workflow Clarity | The 5-step workflow is clearly sequenced (extract → fan out → synthesize → present → handle failure). Explicit validation via the 'If nothing found' step prevents fabrication. Mandatory timeouts serve as safety checkpoints. The anti-patterns section provides clear guardrails against common failure modes. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear headers and numbered sources, but everything is inline in a single file. The 8 search sources section is fairly long and could benefit from being split into a reference file, with SKILL.md providing just the overview and workflow. However, for a skill of this size (~100 lines), it's borderline acceptable. | 2 / 3 |
Total | 11 / 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 | |
ce9ca8e
Table of Contents
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.