recall
A practical skill that automatically surfaces relevant past learnings at session start, improving productivity by preventing repeated mistakes and leveraging previous discoveries.
Packaged view
This page reorganizes the original catalog entry around fit, installability, and workflow context first. The original raw source lives below.
Install command
npx @skill-hub/cli install saadshahd-moo-md-recall
Repository
Skill path: hope/skills/recall
A practical skill that automatically surfaces relevant past learnings at session start, improving productivity by preventing repeated mistakes and leveraging previous discoveries.
Open repositoryBest for
Primary workflow: Ship Full Stack.
Technical facets: Full Stack.
Target audience: Developers and technical professionals who work on complex projects across multiple sessions and want to maintain context and avoid repeating mistakes..
License: Unknown.
Original source
Catalog source: SkillHub Club.
Repository owner: saadshahd.
This is still a mirrored public skill entry. Review the repository before installing into production workflows.
What it helps with
- Install recall into Claude Code, Codex CLI, Gemini CLI, or OpenCode workflows
- Review https://github.com/saadshahd/moo.md before adding recall to shared team environments
- Use recall for productivity workflows
Works across
Favorites: 0.
Sub-skills: 0.
Aggregator: No.
Original source / Raw SKILL.md
--- name: recall description: Auto-activates at session start to surface relevant learnings. Use when starting work in a domain to recall past insights from ~/.claude/learnings/. --- # Recall Skill Surface relevant learnings from past sessions. ## When This Skill Activates - Session start (new or resumed) - Before substantial work in a domain - When soul skill's Silent Audit prompts "Learnings recalled?" - Explicitly via `/hope:recall [context]` ## Input Optional context hint (e.g., "hooks", "testing", "typescript"). If empty, infer from current project/conversation. ## Process 1. **Read learnings files** using the Read tool: - `~/.claude/learnings/failures.jsonl` - `~/.claude/learnings/discoveries.jsonl` - `~/.claude/learnings/constraints.jsonl` If files don't exist, skip silently. 2. **Filter by relevance**: - Match `context` field against provided hint or inferred domain - Match `applies_to` tags against current work - Prioritize recent entries (last 30 days) - Prioritize high-confidence discoveries (>= 0.8) 3. **Output format**: ### Relevant Failures - **[context]**: [failure] → Prevention: [prevention] ### Relevant Discoveries - **[context]** (confidence: X): [discovery] ### Active Constraints - **[context]**: [constraint] (permanent: yes/no) 4. **If no relevant learnings**: Report "No learnings found for [context]"