Free tools Windows power users keep installed
One-click scans. No signup required.
Yes: changing a helper file can make old test evidence unsuitable even when the skill’s SKILL.md is untouched. TraceMantle helps teams check whether supplied evaluation evidence still matches the skill package they intend to release. It analyzes files and evidence; it does not run the skill or prove how an agent will perform in a live environment.
Why SkillCheck became TraceMantle
Project maintainer Brad Kinnard announced the rename from SkillCheck to TraceMantle on September 13, 2026. He said the old name conflicted with another project and that the new name better described the expanded scope: validating skill files, tracking the wider package, and checking evidence used to approve a version. Kinnard’s announcement explains the change.
As an Amazon Associate I earn from qualifying purchases.
An agent skill is not necessarily just its main SKILL.md. A package may also contain helper scripts, reference documents, templates, and schemas. Those supporting files can affect behavior. If one changes, a report from tests against the previous package may no longer cover the release candidate—even if the instructions themselves are identical.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchHow TraceMantle checks whether evidence still applies
- Validate the skill files. TraceMantle checks items such as
SKILL.mdfrontmatter, file references, size limits, and compatibility advice against the Agent Skills specification. - Track the complete package. It records a manifest of package files and content fingerprints, making changes to supporting resources visible even when
SKILL.mdhas not changed. - Import evaluation evidence. It supports a version-pinned Promptfoo export format and preserves the original export while recording details about evaluated inputs, configuration, checks, and execution context.
- Compare evidence with the release candidate. TraceMantle checks the supplied evidence against the candidate package and an approved policy. It can report changed inputs, missing or incompatible evidence, and checks that should be run again.
- Apply a separately trusted policy. The policy comes from a selected trusted Git revision. The candidate package cannot replace the policy or make required checks optional. A passing result in an imported report is not automatically trusted; the project owner or trusted automation must validate it against the policy.
This separation matters: the candidate is the thing being assessed, not the authority that decides what counts as sufficient evidence.
What pass, fail, and unknown mean
A failed check is different from evidence that is absent or unsuitable. TraceMantle uses unknown when the available evidence does not establish a pass. That status is not a successful release check, but it is also not proof that the skill itself failed. Teams should treat it as unresolved coverage and decide whether to obtain suitable evidence or block approval under their policy.
What TraceMantle does—and does not—prove
TraceMantle analyzes files and supplied evidence; it does not execute the skill or evaluator. A structurally valid package or an imported model judgment therefore cannot establish that an agent will perform a task correctly in a live environment. The project also notes that dependency analysis cannot infer every runtime dependency or arbitrary programming-language import. Trusted checks need accurate declarations of their inputs; incomplete coverage calls for further evaluation.
The project repository and package listing describe TraceMantle as a local Python CLI and library, with PyPI installation, a composite GitHub Action, a pre-commit hook, and a Python API. The listing specifies Python 3.10 or later. Check the project’s current instructions for exact installation and integration commands, since release details can change: TraceMantle on GitHub and TraceMantle on PyPI.
How to interpret the project’s testing claims
The project listing reports 1,407 tests, with CI across Python 3.10 through 3.13 and Linux, macOS, and Windows. These are figures reported by TraceMantle’s maintainers, not independent measurements. They describe the project’s own test and CI coverage; they do not demonstrate live agent performance.
Version details should be read with their dates. Kinnard’s September 13, 2026 article identified two remaining issues: Markdown dependency detection and numeric JSON input handling. The repository and package listing surfaced version 1.6.1, described as correcting dependency discovery across Markdown headings and rejecting numeric overflow in evidence imports. Those are different snapshots; consult the current release record rather than treating the earlier issue status as current.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.When this workflow is useful
- Your skill package includes scripts, schemas, templates, or reference files that may change independently of the instructions.
- You want a release check to consider whether imported evaluation evidence covers the exact candidate bundle.
- You need policy requirements to come from a trusted revision rather than from the candidate being evaluated.
- You want missing or unsuitable evidence represented separately from a failed check.
If the question is whether the agent actually succeeds at a task under real operating conditions, file and evidence analysis alone is not enough; that requires execution or other suitable evaluation beyond what TraceMantle itself performs.
Quick Recap
Best Value
Rank #4
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
