Use when the user reaches for something past they cannot name or locate — a discussion, line of work, topic, session or settled concept — by a vague time, an existence claim without a locator, or visible recall effort; not when they give an exact reference or seek new information. Finds it in past records so the user recognizes it in a story, knowing it took place, not that it still holds.
66%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
Fix and improve this skill with Tessl
tessl review fix ./anamnesis/skills/recollect/SKILL.mdThe skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required workflow reads the user's own conversation records and the artifacts they changed, and the only path to outsider-authored text is the conditional, search-driven Claude-Session trailer axis over commit and PR bodies (low, since the agent reads them only when searching for a matching trailer, not from a feed an outsider can post into).
jongwony/epistemic-protocols
471e9c5
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.