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 bledden/claude-recall-plugin --skill recall-assistantgit clone --depth 1 https://github.com/bledden/claude-recall-pluginWrote 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/bledden/claude-recall-plugin/recall-assistant)<a href="https://agentmods.dev/skills/bledden/claude-recall-plugin/recall-assistant"><img src="https://agentmods.dev/badge/skills/bledden/claude-recall-plugin/recall-assistant.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 190 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00056 | $0.01755 |
| Opus 5 | $0.00028 | $0.00877 |
| Sonnet 5 | $0.00011 | $0.00351 |
| Haiku 4.5 | $0.00006 | $0.00176 |
Grade A, and why
recall-assistant 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 — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recall Assistant
You have access to the recall plugin — a cross-session, cross-project conversation recall system backed by SQLite. This skill guides you on when and how to use it proactively.
IMPORTANT: All behaviors below are gated. Before taking any proactive action described here, check whether the user has enabled this skill by verifying skill_enabled is true. If not enabled, do nothing — the user interacts with the plugin only via explicit /recall commands.
Context-Loss Detection
When skill_enabled is true, watch for signals that you have lost earlier conversation context. The user may configure which signal categories are active via detection_signals (default: all three).
Explicit Signals
Hook-backed (reliable): when
skill_enabledis true, theUserPromptSubmithook already detects these explicit phrases deterministically and injects a[Recall]suggestion into your context as a system reminder — so this signal no longer depends on you noticing. When you see that reminder, act on it (run the appropriate recall). The behavioral and temporal signals below remain model-driven.
Watch for the user saying things like:
- "didn't we already discuss..."
- "what was that thing about..."
- "earlier you said..."
- "I think we talked about..."
- "we discussed this before"
- "remind me what..."
- "you mentioned something about..."
When you detect these: suggest a specific recall command based on what the user is looking for.
Example:
I may have lost context on the warp divergence discussion. Let me check — I'll run
/recall search "warp divergence"to recover what we covered.
Behavioral Signals
Watch for your own behavior indicating context loss:
- You are about to ask a question that the user likely already answered earlier in the session
- You are repeating advice or explanations you gave before without realizing
- You are about to contradict something you said earlier
When you detect these: pause, acknowledge the potential gap, and suggest a recall command before continuing.
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 Changed · +1 lines f65eb51d0bf2
- 8d ago First seen · 193 lines · 56 tokens per session scan A d92274b21c36
recall-assistant is a skill published in the GitHub repository bledden/claude-recall-plugin (23 stars, last pushed 2d ago), licensed MIT. It adds 56 tokens to every session and 1,755 once invoked, about $0.0003 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
sitrep
Situational awareness — where am I, what was I doing, what's next. Context recovery after compression, confusion, or mid-session reorientation.
ambient-context-catch-me-up
Use when the user asks what they were doing, where they got to, what is still open, or to be caught up after a break. Reads today's Ambient Context summary and record and briefs them.
engineer-brain
Your self-updating engineering brain — tracks work patterns, expertise, habits, progress gaps, and generates daily sync-up notes and quarterly review content. Invoke when you say "engineer brain", "update brain", "daily sync", "standup", "sync up", "quarterly connections", "what did I work on", "my progress", "where…
AgentMail
Give the agent its own dedicated email inbox via AgentMail. Send, receive, and manage email autonomously using agent-owned email addresses (e.g. [email protected]).
Apple Notes
Open Notes.app and create or update iCloud notes on macOS using memo when possible and AppleScript when direct app automation is more reliable.
Himalaya Email
Work with IMAP email from the terminal using Himalaya for inbox reads, draft review, and carefully confirmed outbound sends.