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 ntorga/agent-starter-kit --skill contextualizer-self-reviewgit clone --depth 1 https://github.com/ntorga/agent-starter-kitWrote 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/ntorga/agent-starter-kit/contextualizer-self-review)<a href="https://agentmods.dev/skills/ntorga/agent-starter-kit/contextualizer-self-review"><img src="https://agentmods.dev/badge/skills/ntorga/agent-starter-kit/contextualizer-self-review/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/ntorga/agent-starter-kit/contextualizer-self-review"><img src="https://agentmods.dev/badge/skills/ntorga/agent-starter-kit/contextualizer-self-review.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 | $0.00026 | $0.02129 |
| Opus 5 | $0.00013 | $0.01064 |
| Sonnet 5 | $0.00005 | $0.00426 |
| Haiku 4.5 | $0.00003 | $0.00213 |
Grade A, and why
contextualizer-self-review 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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
Before delivering context files or briefs, the Contextualizer evaluates its own output against the TRACE rubric. Each letter is scored 0, 1, or 2 with evidence quoted from the rubric and cited from actual work. The total determines whether to deliver, rewrite, or abort.
Procedure
-
Gather evidence. Honest self-review makes verification efficient. Accurate scorecards confirm fast; dishonest ones fail and re-run, wasting time and compute. Before scoring, run verification commands to gather proof. The examples below show common patterns — choose what provides the best evidence for your specific work.
Examples:
- Schema compliance: verify
.context.mdfiles have opening<context>tag with path and date, Summary, Constraints, Guidance sections - Feature map: verify
FEATURE-MAP.mdentries have feature name, flow steps with file paths and role descriptions - Anchored claims:
test -f <path>for directories/files mentioned in context — confirms claims are grounded - Coverage:
find . -type f | wc -landfind . -type d | wc -lto check project size against yield threshold - Brevity:
wc -l <context-files>— each context file should be shorter than the directory it describes
These are examples, not mandates. Choose commands that provide the strongest proof for your output.
- Schema compliance: verify
-
Score each criterion. Read the TRACE rubric below. For each letter, assign a score of 0, 1, or 2. You must:
- Quote the specific criterion level (0, 1, or 2) that your work matches.
- Cite evidence from your actual work — files verified, schema elements checked, line counts measured. Generic claims like "I verified everything" are not evidence and score 0.
-
Output the Scorecard. Fill in the scorecard below. This is not internal reasoning — this is your deliverable checkpoint.
-
Apply the hard-fail rule. If any letter scores 0, do not deliver — go to step 5 immediately.
-
Determine action by total score:
- 9 – 10 — DELIVER — Output meets all criteria. Deliver to user.
- 7 – 8 — FIX the scored < 2 criteria. a. Identify which letters scored below 2. b. Fix those gaps automatically (do NOT consult the user). c. Re-score, then deliver if 9-10. d. If still below 9, retry once more. e. After 2 failed fix attempts, yield with the current state, rubric scores, and blocking letters.
- 0 – 6 — RESTART — The output is fundamentally broken. Rewrite from scratch with corrected understanding, or yield to the user with an explanation of what went wrong.
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 First seen · 109 lines · 26 tokens per session scan A aa92561c1779
contextualizer-self-review is a skill published in the GitHub repository ntorga/agent-starter-kit (142 stars, last pushed 2d ago), licensed MIT. It adds 26 tokens to every session and 2,129 once invoked, about $0.0001 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 2026-09-13.
Other skills, from other repositories
agent-framework-py-release
Use when cutting a Python release for the microsoft/agent-framework monorepo. Triggers on "bump py versions", "cut a python release", "prepare release PR for python", "release py packages", "bump python to X.Y.Z", or similar requests to bump Python package versions and prepare a release PR. Handles all four lifecycle…
python-package-management
Guide for managing packages in the Agent Framework Python monorepo, including creating new connector packages, versioning, and the lazy-loading pattern. Use this when adding, modifying, or releasing packages.
foundry-hosted-agent-validation
Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.
verify-samples-tool
How to use the verify-samples tool to run, verify, and manage sample definitions in the Agent Framework repository. Use this when adding, updating, or running sample verification.
build-and-test
How to build and test .NET projects in the Agent Framework repository. Use this when verifying or testing changes.
python-feature-lifecycle
Guidance for package and feature lifecycle in the Agent Framework Python codebase, including stage meanings, feature-stage decorators, feature enums, and how to move APIs from one stage to the next.