Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/cofin/flow/flow-memory-keepernpx skills add cofin/flow --skill flow-memory-keepergit clone --depth 1 https://github.com/cofin/flowWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00048 | $0.01124 |
| Opus 5 | $0.00024 | $0.00562 |
| Sonnet 5 | $0.00010 | $0.00225 |
| Haiku 4.5 | $0.00005 | $0.00112 |
Grade A, and why
flow-memory-keeper scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Flow Memory Keeper
Purpose
This project-local skill prevents Flow work from drifting into stale specs, missing learnings, or half-archived folders.
Use it whenever work is completed, paused, blocked, revised, synced, or archived.
Mandatory Outcomes
Before claiming a task, phase, or flow is complete:
- Ensure
spec.mdis readable and current through the normal Flow sync process. - Capture concrete learnings in the flow's
learnings.md. - Capture failures, false starts, blockers, and recovery notes when they would help a future session.
- If the user had to repeat a correction or showed frustration that something was forgotten, flag that as a workflow gap and capture it explicitly.
- Elevate durable patterns to evidence-backed
.agents/bundles/knowledge/patterns/<topic>.mdchapters. - Update chapters recursively under
.agents/bundles/knowledge/when the current-state knowledge base changed, preserving project-shaped relative paths, indexes, and links. - If the flow is complete, archive it cleanly and leave the active spec area uncluttered.
- Capture validated repo-native commands and verification workflows in
.agents/bundles/knowledge/workflow.mdwhen they were discovered or corrected during the work.
Completion Protocol
Task or Phase Completion
- Append a concise entry to
.agents/bundles/specs/<flow_id>/learnings.md:- what changed
- why it changed
- files touched
- commands or checks that mattered
- canonical repo commands that future agents should reuse
- gotchas, failures, and recoveries worth remembering
- any repeated user correction or frustration that revealed a missing default, checklist item, or workflow rule
- Move reusable guidance into a topic-specific
.agents/bundles/knowledge/patterns/<topic>.mdchapter. - If the work changed architecture, conventions, tooling, operational behavior, or canonical project commands, recursively locate and update the relevant chapter under
.agents/bundles/knowledge/at its existing project-shaped relative path; keep.agents/bundles/knowledge/workflow.mdas the stable command default. - Promote repeated user corrections or frustration into an obvious durable rule instead of leaving it as a one-off note.
- Run the normal Flow sync step so
spec.mdreflects the latest state.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 118 lines · 48 tokens per session scan A 3d440cc39fdd
flow-memory-keeper is a skill published in the GitHub repository cofin/flow (15 stars, last pushed 3d ago), licensed Apache-2.0. It adds 48 tokens to every session and 1,124 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
lavra-knowledge
Capture solved problems as knowledge entries for fast recall. Use when a solution should be preserved for future sessions.
creating-agent-skills
Create and refine Claude Code Skills. Use when authoring SKILL.md files, building new skills, or improving existing skill structure and best practices.
compound
Extract lessons, patterns, and architectural truths to compound team knowledge.
openspec-onboard
Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
a-evolve
Apply A-Evolve's agentic evolution methodology to improve AI agent performance across runs. Use when the user wants to diagnose agent failures, generate targeted skills from error patterns, evolve system prompts, or accumulate episodic knowledge. Works standalone or inside AutoResearchClaw pipelines. Triggers on…
code-documenter
Use when adding docstrings, creating API documentation, or building documentation sites. Invoke for OpenAPI/Swagger specs, JSDoc, doc portals, tutorials, user guides.