recall

recall is a skill for Claude Code, Codex from mostlyharmless-ai/watercooler. It costs 41 tokens per session (532 once invoked), scanned A, original, Apache-2.0.

A project-memory tool for finding earlier context, decisions, and reasons in an unfamiliar codebase. It can answer questions about project history or summarize relevant past discussions.

In plain words
What is it for?
Use it to recall decisions, implementation patterns, known issues, and related discussions about the current repository.
Why use it?
It prevents repeated decisions and reduces the time spent searching old project information. It is useful before starting work or investigating why an implementation was chosen.

Skill for Claude CodeCodex

Part of the watercooler plugin — 7 skills, 1 MCP server shipped together

Install

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.

agentmods
npx agentmods add skills/mostlyharmless-ai/watercooler/recall
Any agent
npx skills add mostlyharmless-ai/watercooler --skill recall
Clone the repo
git clone --depth 1 https://github.com/mostlyharmless-ai/watercooler

Made for: Claude Code, Codex.

Or install watercooler, the plugin that ships this one along with the rest of its 7 skills, 1 MCP server.

Wrote 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.

agentmods badge for recall

README.md
[![agentmods](https://agentmods.dev/badge/skills/mostlyharmless-ai/watercooler/recall.svg)](https://agentmods.dev/skills/mostlyharmless-ai/watercooler/recall)
Your own site
<a href="https://agentmods.dev/skills/mostlyharmless-ai/watercooler/recall"><img src="https://agentmods.dev/badge/skills/mostlyharmless-ai/watercooler/recall.svg" alt="Measured on agentmods" height="20"></a>
Per session 41 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 532 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00041 $0.00532
Opus 5 $0.00020 $0.00266
Sonnet 5 $0.00008 $0.00106
Haiku 4.5 $0.00004 $0.00053

Measured 5d ago against content hash 6caef11d56b8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

recall 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 5d 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.

plugins/claude/watercooler/skills/recall/SKILL.md · 63 lines

What it actually says

Recall

Query: $ARGUMENTS

Framing Detection

After stripping flags (e.g. --raw), inspect the remaining text:

  • Question framing if text ends with ? OR starts with a question word: What, Why, How, Who, When, Where, Which, Is, Are, Was, Were, Did, Does, Can, Should
  • Context framing otherwise (topic or task description)

Steps

  1. Load MCP tool:

    ToolSearch: select:mcp__watercooler__watercooler_smart_query
    
  2. Execute query (scope to current repo with code_path):

    mcp__watercooler__watercooler_smart_query(query="<$ARGUMENTS minus flags>", code_path="<repo root>")
    
  3. Present results based on framing:

    Question framing — answer the question:

    • Direct answer to the question
    • Evidence — cite specific entries by entry_id
    • Confidence and tier used (T1/T2/T3)
    • Escalation note if tier escalated beyond T1
    • Suggested follow-ups if answer is incomplete or uncertain

    Context framing — summarize relevant context:

    • Prior decisions related to this topic
    • Relevant patterns or implementations
    • Known issues or gotchas
    • Related threads for deeper reading
    • Tier used and escalation reason
  4. Handle empty results:

    • Suggest alternative search terms
    • Note whether memory backends are configured
  5. Raw output (if --raw flag present):

    • Append full JSON response after the summary

Example Invocations

  • /recall authentication flow — context before implementing auth
  • /recall What was decided about the config system? — direct question
  • /recall Why did we choose markdown for threads? — direct question
  • /recall --raw branch parity sync — raw JSON output
  • /recall Who implemented the branch parity feature? — attribution lookup
Changes

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.

  1. 5d ago First seen · 63 lines · 41 tokens per session scan A 6caef11d56b8

Subscribe to this mod's changes

recall is a skill published in the GitHub repository mostlyharmless-ai/watercooler (13 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 41 tokens to every session and 532 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.