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 skills add zircote/subcog --skill memory-capturegit clone --depth 1 https://github.com/zircote/subcogWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zircote/subcog/memory-capture)<a href="https://agentmods.dev/skills/zircote/subcog/memory-capture"><img src="https://agentmods.dev/badge/skills/zircote/subcog/memory-capture/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zircote/subcog/memory-capture"><img src="https://agentmods.dev/badge/skills/zircote/subcog/memory-capture.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What 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 5.1 | $0.00000 | $0.00798 |
| Opus 5.5 | $0.00000 | $0.00319 |
| Sonnet 5.5 | $0.00000 | $0.00160 |
| Haiku 4.5 | $0.00000 | $0.00080 |
Grade A, and why
memory-capture 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 1mo 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memory Capture Skill
Capture decisions, learnings, patterns, and context as persistent memories that survive across sessions.
Trigger Phrases
- "remember this", "capture this", "save this decision"
- "document that", "log this pattern", "note this learning"
- "TIL", "gotcha", "we decided", "going with"
- "this is important", "don't forget"
Quick Reference
| Namespace | Use When | Signal Words |
|---|---|---|
decisions |
Making architectural or design choices | "decided", "chose", "going with" |
patterns |
Establishing recurring practices | "always", "never", "when X do Y" |
learnings |
Discovering something new | "TIL", "gotcha", "discovered" |
context |
Recording background information | "because", "constraint", "requirement" |
tech-debt |
Noting future work | "TODO", "FIXME", "temporary" |
apis |
Documenting API contracts | "endpoint", "API", "schema" |
config |
Recording configuration details | "config", "environment", "setting" |
security |
Security-related information | "auth", "vulnerability", "permission" |
performance |
Performance insights | "optimization", "benchmark", "latency" |
testing |
Testing strategies | "edge case", "fixture", "coverage" |
Execution Strategy
Capture Quality Guidelines:
- Include context and rationale, not just the decision
- Add relevant tags for discoverability
- Reference related files or components
- Keep content concise but complete
Interactive Capture Workflow
- Detect capture signals in user's message
- Identify the namespace from context and signal words
- Extract the core insight - what should be remembered?
- Enhance with context - why is this important?
- Suggest tags based on content and current work
- Confirm with user using AskUserQuestion if uncertain
- Execute capture via MCP tool or CLI
- Verify success and report the memory ID
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.
- 1mo ago First seen · 99 lines · 0 tokens per session scan A 55dd3753fca7
memory-capture is a skill published in the GitHub repository zircote/subcog (29 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 798 tokens. 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
proactive-agent
Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve. Now with WAL Protocol, Working Buffer, Autonomous Crons, and battle-tested patterns. Part of the Hal Stack 🦞.
memory-recall
Persistent cross-session memory system with fact extraction, semantic search, and compiled memory layers. Your agent remembers everything across sessions — user preferences, project context, past decisions, personal details. Use when: (1) User references something from a previous conversation, (2) Agent needs project…
clawmem
ClawMem operational reference for agents at query time — the 3-rule escalation gate, MCP tool routing, the 4 query-optimization levers, pipeline behavior (query vs intentsearch), composite scoring, and memory lifecycle (pin/snooze/forget). Use when tuning retrieval, troubleshooting recall quality, or any ClawMem…
cocosearch-review-pr
Use when reviewing a GitHub PR or GitLab MR by URL. Fetches diff and metadata via API, then uses CocoSearch for blast radius analysis, dependency impact, pattern consistency, and test coverage assessment. Read-only by default; optionally pushes findings back as inline PR/MR comments after your confirmation.
cocosearch-add-language
Use when adding support for a new programming language or config format to CocoSearch. Guides through all paths (handler, symbol extraction, context expansion) with registration checklists. For grammar handlers, use cocosearch-add-grammar.
cocosearch-add-extractor
Use when adding a dependency extractor for a language or grammar. Guides through pre-checks, extractor implementation, optional module resolver, tests, and registration. Enables deps tree, deps impact, and dependency-enriched search for the target language.