Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add naimkatiman/continuous-improvement/plugin install continuous-improvementWrote 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/naimkatiman/continuous-improvement/proceed-with-the-recommendation)<a href="https://agentmods.dev/skills/naimkatiman/continuous-improvement/proceed-with-the-recommendation"><img src="https://agentmods.dev/badge/skills/naimkatiman/continuous-improvement/proceed-with-the-recommendation/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/naimkatiman/continuous-improvement/proceed-with-the-recommendation"><img src="https://agentmods.dev/badge/skills/naimkatiman/continuous-improvement/proceed-with-the-recommendation.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00083 | $0.10225 |
| Opus 5 | $0.00042 | $0.05112 |
| Sonnet 5 | $0.00017 | $0.02045 |
| Haiku 4.5 | $0.00008 | $0.01022 |
Grade C, and why
proceed-with-the-recommendation scanned grade C with 3 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
Edit `routes/users.ts` adding zod schema. Run `curl -X POST .../api/users -d '{}'` → returns `400 {"errors":[{"field":"email","msg":"required"}]}`. ✓ Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
| Settings / hooks / permission change | `update-config` | Edit `~/.claude/settings.json` with a minimal patch; restart session. (Reference behavior — does not require `update-config`.) | Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- The item is fully verifiable with a single smallest check (one typecheck, one curl, one read-back, one command output snippet). How it starts
The opening of the file, as written. The whole thing — 525 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Proceed With The Recommendation
Overview
One reusable skill that ties the continuous-improvement framework to a concrete execution engine. When the user greenlights a list of recommendations, this skill walks them in order under the 7 Laws, routes each item to the best companion (workspace-surface-audit, superpowers:*, ralph, continuous-improvement helpers), and falls back to explicit inline behavior when a companion is not installed. Never silent no-op, never drive-by, never skip verification.
Core principle: execute in order, one concern at a time, verify before advancing, reflect at the end.
The 7 Laws (inline summary)
This skill is the execution engine for the continuous-improvement framework. Mirrored from continuous-improvement/SKILL.md so this file stands alone if that companion is not installed:
- Research Before Executing — what exists, what constrains, what breaks, simplest path. If you can't answer, research first.
- Plan Is Sacred — state WILL build / Will NOT build / Verification / Fallback before any edit.
- One Thing at a Time — finish and verify one item before the next. No "also quickly add" drive-bys. Parallel only when items share no state.
- Verify Before Reporting — "Done" requires actual output checked, not assumed. "Should work" is not verification.
- Reflect After Every Session — what worked / what failed / what you'd do differently / rule to add / next 3 ranked moves.
- Iterate Means One Thing — one change → verify → next. Never bundle untested edits.
- Learn From Every Session — capture patterns as instincts; corrections weaken, confirmations strengthen.
The Loop: Research → Plan → Execute → Verify → Reflect → Learn → Iterate. Skipped step = the step you needed most.
Composition Map
This skill is the orchestrator for the other companions in this repo. It does not duplicate them — it calls them.
| Phase | Companion invoked | 7 Laws alignment |
|---|---|---|
| Pre-flight environment check (optional, first-time or big session) | workspace-surface-audit |
Law 1: Research Before Executing |
| Plan restatement + tagging + routing decision | superpowers:writing-plans (if plan > 3 items or > 150 LOC) else inline |
Law 2: Plan Is Sacred |
| Per-item specialist routing | superpowers:* / simplify / schedule / loop / security-review / documentation-lookup / update-config / commit-commands:* (see routing table) |
Law 3: One Thing at a Time |
| Per-item verification | superpowers:verification-before-completion or inline smallest-check |
Law 4: Verify Before Reporting |
| Long-running or autonomous execution (PRD-style, multi-iteration) | ralph |
Laws 3 + 6 at scale |
| End-of-run reflection | Write a Reflection block + append to ~/.claude/instincts/<hash>/observations.jsonl |
Laws 5 + 7: Reflect, Learn |
| Iterate if user course-corrects | Restart from Phase 2 with the corrected list | Law 6: Iterate Means One Thing |
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 37f325a860e7
- 10d ago First seen · 525 lines · 83 tokens per session scan C a449fe8f9dd4
proceed-with-the-recommendation is a skill published in the GitHub repository naimkatiman/continuous-improvement (7 stars, last pushed 3d ago), licensed MIT. It adds 83 tokens to every session and 10,225 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 3 findings (sends data to an external url, reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
opensrc
Fetch dependency source code to give AI agents deeper implementation context. Use when the agent needs to understand how a library works internally, read source code for a package, fetch implementation details for a dependency, or explore how an npm/PyPI/crates.io package is built. Triggers include "fetch source for"…
nopus-configure
Configure nopus complexity sensitivity, extra-simple rewrites, rewrite evidence, and Pi response hiding for this user.
nopus-simplify
Rewrite the immediately preceding assistant response with clearer and more direct prose when the user invokes this skill.
agentpush
Bridge an imported agentpush MCP server (Telegram, WhatsApp, etc.) with live agentproto sessions via the daemon's transmitter subsystem: transmitmessage sends outbound and binds the recipient to a session, inboundwatcherstart polls agentpush for new messages, and inboundendpointcreate/POST /inbound(/:slug) route…
bureau
Drive Bureau — the browser stack's installable capability server: a stealth Firefox (Camofox) + daemon that exposes browser tools as MCP-over-HTTP on :8830, plus a CLI for saved browser identities (sessions), social capture / search, adapter health probes, declarative workflows, and Guilde connection. Use when working…
agentproto-apps
Operate and build Agentproto apps — AIP-42 app bundles (defineApp().emit(dir), APP.md + agents/ + workflows/ + ui/) and their daemon lifecycle (appinstall, appapply, apprun, appstatus, appstop) plus the app-scoped durable data plane (appdataread/write/list/migrate). Covers serving one with a UI (agentproto app serve →…