AutoSaddler is a system that improves LLM-agent harnesses by diagnosing execution traces and applying structured changes to prompts, tools, middleware, and agent-loop logic. It evaluates candidate updates for their ability to generalize beyond the traces that motivated them. The catalogue add-ons represent workflows for using AutoSaddler.
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 microsoft/AutoSaddler --skill verification-baselinegit clone --depth 1 https://github.com/microsoft/AutoSaddlerWrote 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/microsoft/autosaddler/verification-baseline)<a href="https://agentmods.dev/skills/microsoft/autosaddler/verification-baseline"><img src="https://agentmods.dev/badge/skills/microsoft/autosaddler/verification-baseline/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/microsoft/autosaddler/verification-baseline"><img src="https://agentmods.dev/badge/skills/microsoft/autosaddler/verification-baseline.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00028 | $0.00450 |
| Opus 5.5 | $0.00011 | $0.00180 |
| Sonnet 5.5 | $0.00006 | $0.00090 |
| Haiku 4.5 | $0.00003 | $0.00045 |
Grade A, and why
verification-baseline 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 1mo 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.
What it actually says
Verification Baseline (Core)
Purpose (Core)
Verification asks whether the complete candidate is internally consistent, loadable, within scope, and capable of producing the intended effect without obvious collateral damage. It is not permission to weaken the intended change until a check passes.
Procedure (Core)
Before returning a change:
- Review the complete diff or update set. Remove unintended edits, debug text, stale artifacts, no-op churn, and unrelated formatting.
- Confirm every changed unit is inside the declared mutation scope, every declared unit actually changed, and every protected input is untouched.
- Trace dependencies and callers of shared surfaces. Preserve registration, imports, defaults, signatures, schemas, side effects, and companion files unless the diagnosis requires a coordinated change.
- Validate syntax and structure with the plugin's available checks. Confirm the candidate loads through its real integration path, not only an isolated parser.
- Inspect the final rendered or agent-visible form of prompts, descriptions, templates, hooks, and schemas. Check placeholders, escaping, formatting, examples, parameter names, and runtime behavior for consistency.
- Exercise empty inputs, boundary values, error paths, and representative existing callers when they can reach the changed surface.
- Recheck behavior that already passed and could be affected. A targeted fix should not silently broaden, suppress, or reorder unrelated behavior.
- Confirm the mutation is substantive: the effective candidate differs in the intended way after normalization, rendering, or serialization.
- Return only when the proposed effect, actual changed units, verification evidence, and structured result agree exactly.
Failure Handling (Core)
If verification fails, repair or remove the invalid change and rerun the same check. Do not hide the failure, modify protected evaluation logic, fabricate a successful result, or change intended semantics merely to satisfy a superficial check. The plugin-specific verification skill may impose stricter rules; those rules are part of the candidate contract.
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.
- 1mo ago First seen · 47 lines · 28 tokens per session scan A faef9229ba28
verification-baseline is a skill published in the GitHub repository microsoft/AutoSaddler (243 stars, last pushed 3d ago), licensed MIT. It adds 28 tokens to every session and 450 once invoked, about $0.0001 per session on Opus 5.5. 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
adk-go-self-review
Review an ADK Go change the way a maintainer will — a fresh-context pass over the whole diff, five lenses (correctness and tests, scope, simplicity, style, adk-python parity), and the mutation check that proves your tests pin the change. Use before opening a PR, before any later push that changes code, and when asked…
adversarial-reviewer
Stress-test a code change for concrete correctness defects, unsafe assumptions, and failure modes. Use when the user explicitly asks for an adversarial review, a deep bug hunt, or to tear a change apart, or to check your own change before submitting it. Do not use for ordinary review, style feedback, or general…
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
semgrep-rule-variant-creator
Creates language variants of existing Semgrep rules. Use when porting a Semgrep rule to specified target languages. Takes an existing rule and target languages as input, produces independent rule+test directories for each language.
brooks-sweep
Full-sweep mode: runs a unified analysis across all quality dimensions — code decay, architecture, tech debt, and test quality — then applies fixes directly to the codebase. Safe changes are auto-applied; risky changes are confirmed before execution. Drawing on twelve classic engineering books. Triggers when: user…
plugin-review
Review plugin quality with tiered checks and dependency scoping. Use for PR and pre-release audits.