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/acn-ericlaw/agent-memory/second-opinionnpx skills add acn-ericlaw/agent-memory --skill second-opiniongit clone --depth 1 https://github.com/acn-ericlaw/agent-memoryWrote 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/acn-ericlaw/agent-memory/second-opinion)<a href="https://agentmods.dev/skills/acn-ericlaw/agent-memory/second-opinion"><img src="https://agentmods.dev/badge/skills/acn-ericlaw/agent-memory/second-opinion.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00066 | $0.01974 |
| Opus 5 | $0.00033 | $0.00987 |
| Sonnet 5 | $0.00013 | $0.00395 |
| Haiku 4.5 | $0.00007 | $0.00197 |
Grade A, and why
second-opinion 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 4d 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
second-opinion
⚠️ Tool-managed skill provided by agent-memory (
provenance: agent-memory-builtin). Don't edit it in place — it is overwritten on upgrade. To change behavior: fork it under a new skill name, or upstream a genuine fix to the agent-memory project (file an issue in its repo, or bring it to the tool maintainer) for back-port + validation. SeeSKILLS.md→ "Tool-provided (system) skills".
A deliberate fresh-context review ritual. A long session accumulates assumptions and self-trust — the agent that built a solution over-trusts its own trajectory. This skill packages a compact snapshot of the current task so a reviewer with clean memory (a fresh session, or a different vendor) can challenge it. The reviewer's value is precisely that it did not live the session.
Pairs with apply-critique, which consumes the critique the reviewer hands back.
When to use
- Milestone mode — the work looks complete and you're about to close it. This is the highest-value case: the in-session agent is least likely to challenge itself exactly when it feels done.
- Reactive mode — you're blocked, uncertain, the logic seems inconsistent, several designs compete, or the change is high-stakes.
Invoked by the human, on demand. It never runs during ordinary work.
What to do
-
Determine the trigger from how the human invoked you —
milestonevs.reactive. Note it in the snapshot's Exchange Intent. -
Show the security advisory and wait for explicit acknowledgment. Packaging task state for another AI system is a trust-boundary event, even when the reviewer is internal. Print this verbatim and stop until the human types
acknowledge(or cancels) — do not generate the snapshot first:━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Fresh-Context Review — Snapshot Advisory ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ You are about to generate a work-in-progress snapshot for review by another AI system (a clean session or a different vendor). Before you share it, ensure: - no client secrets, credentials, or tokens are included; - no personally identifiable information (PII) is exposed; - the content is appropriate to share beyond this project's perimeter. You may prefer a reviewer inside your organization's security boundary. Even then, review before sharing — think of it as handing context to a trusted collaborator. Power comes with responsibility. Type "acknowledge" to proceed, or cancel. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
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.
- 4d ago First seen · 175 lines · 66 tokens per session scan A 5a0ff4b490f1
second-opinion is a skill published in the GitHub repository acn-ericlaw/agent-memory (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 66 tokens to every session and 1,974 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 2026-08-31.
Other skills, from other repositories
reelier-slice
Use when implementing any change to the Reelier authority ledger or Path C compiled-authority code - encodes the evidence-led slice loop, the baseline gate, the RED/GREEN review protocol, and the failure modes that have actually cost slices here.
reelier-replay
Freeze a repeatable, tool-call-driven task into a replayable Reelier skill, and replay an existing skill instead of redoing deterministic work. Use after finishing a task that was mostly API/MCP tool calls (data pulls, report generation, deploy checks, CRUD sequences) — never for arbitrary coding/file-edit sessions…
reelier-write-safety
Bound and record an agent's writes before granting them. Use when the user is adding or configuring an MCP server, planning to run an agent unattended or on a schedule or in CI, or deciding whether to give an agent write access to a real system. Covers the recorder, the policy seatbelt, approvals, and what a receipt…
reelier
Replay recurring deterministic jobs at 0 tokens instead of re-reasoning them every run. For cron/heartbeat tool-call workflows (status checks, data pulls, report generation, CRUD sequences) — record the job once with reelier, then run "npx -y reelier run " on every heartbeat and gate drift with "reelier diff". Never…
writing
将共享历史中的已验证事实和计算结果整理成符合受众、格式与长度约束的成稿。.
kayba-stage-7-fixer
Implement the approved fixes from the action plan and log all changes. Trigger when the user says "run stage 7", "implement fixes", "apply action plan", or when invoked by the kayba-pipeline orchestrator. Requires eval/actionplan.md to exist.