Rules File Certified

liatrio-agents-md

Splits large or mixed-concern Git branches into smaller, reviewable pull request stacks with safety references and merge sequencing.

68

/ 100 · Grade D

D = 60–69

I need to break down oversized or mixed-concern branches into smaller, manageable pull requests that are easier to review and less prone to conflicts.

gitversion controlcode reviewworkflowdevelopment tools
Publisher: liatrio-labsVersion: latestCertified: Mar 28, 2026Expires: Mar 28, 2027Source ↗

liatrio-agents-md earned Certified status with a trust score of 68/100 (Grade D). Adversarial testing produced 1 finding (1 high). Security scan flagged 0 findings.


Trust Score Breakdown

Eight weighted signals composing the aggregate trust score

security scan
100% × 15w
15.0
supply chain
100% × 10w
10.0
adversarial
98% × 25w
24.5
provenance
80% × 20w
16.0
consumer confirm
0% × 10w
0.0
behavioral pass
0% × 10w
0.0
contract accuracy
100% × 6w
6.0
uptime
100% × 4w
4.0

Scheme v2.0 · Weights provisional · Consumer confirmations and uptime use pipeline-derived baselines.


Findings

Security scan results, adversarial testing, and pipeline review

Security Scan — Cisco Skill Scanner

cisco-skill-scannerFAIL
staticbehavioralllmmeta
critical: 0high: 0medium: 0low: 0

Adversarial Testing — 4 categories, 1 findings

prompt injection chainscapability squattingcontext poisoningdependency confusion
highReviewprompt injection chainssuspected

Several skills reference user-provided content (files, messages, data) without explicit data-boundary markers. The git-commit-conventional skill processes git diffs and user requests, the create-pull-request skill analyzes branch changes, and the mastra-api skill processes agent IDs and workflow names from user input. While these skills appear to treat user content as data within their scope, the lack of explicit data boundaries could potentially allow instruction-like user content to influence skill behavior.

Methodology v1.0 · 4 categories · ~55 attack patterns


Behavioral Fingerprint

Runtime performance baseline for drift detection

Samples

8

Error rate

0.0%

Peak memory

— MB

Avg CPU

—%

Response time distribution

p50: 7934msp95: 18668msp99: 18668ms

Output size distribution

p50: 1.2 KBp95: 3.9 KBmean: 1.8 KB

Fingerprint v1.0 · Baseline: Mar 28, 2026 · Status: baseline


Interface

Skill triggers and instruction summary

Activation

This skill activates when a PR or branch is too large, difficult to review, mixed across concerns, conflict-prone, or needs to be decomposed without losing net changes.

This skill handles splitting oversized or mixed-concern branches into smaller, reviewable PR stacks with safety refs, topology selection, parity audits, and merge sequencing.

Instructions: 81Files: 53Format: markdown

Does

Quantify branch shape by analyzing commit count, file changes, and dependency density

Present top two topology options with explicit tradeoffs before proceeding

Collaborate with user to select topology using facilitated discovery questions

Create backup refs (tags and/or backup branches) before executing surgery

Execute branch surgery with strict scope boundaries and safety measures

Run mandatory audit gates to verify completeness and parity

Generate PR metadata and merge sequencing documentation

Use cherry-pick with staging for mixed commits to preserve scope contracts

Does not

Execute branch surgery without user collaboration on topology selection

Skip backup ref creation before making changes

Proceed without running audit gates

Allow hidden carryover changes between split branches

Skip tradeoff analysis when presenting topology options


Scope & Permissions

What this capability can and cannot access — derived from pipeline analysis

creates files

yes

deletes files

no

modifies files

yes

accesses env variables

no

invokes external tools

yes

makes network requests

no


Known Failure Modes

Documented edge cases and recovery behaviors

when when audit gates fail

then the agent fixes scope drift or ownership errors and re-runs audits

when when signals are incomplete

then the agent gathers more evidence before recommending topology

when when user is unsure about topology

then the agent recommends one option and explains what would change with alternatives


Badge & Integration

Embed certification status in your README, docs, or CI pipeline

Fidensa Certified badge for liatrio-agents-md
badge SVG →attestation API →integration guide →

Certification Notes

Provenance observations from the pipeline

publisher

Publisher "liatrio-labs" is not verified — first certification from this publisher

provenance

Single contributor — no peer review evidence in commit history

provenance

Repository is 19 days old — recently created

provenance

Package description appears to be boilerplate or template text


Signed Artifact

Certification provenance and verification metadata

Content hashsha256:3f9d5b86eb9d12be9b91ab3da5394f40aa5fd40f530ea8ab8829eae0dae8bd59
Key IDkms-9db4ed3b9f53
CertifiedMar 28, 2026
ExpiresMar 28, 2027
Pipeline version1.0
Statusvalid