Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Fritzandfriends/Semantic Pattern Guardrails — measured AI agent skill | DecimalAI

fritzandfriends/semantic-pattern-guardrails

NewSkillScoreMaintained

Guardrails for implementing Web Forms semantic page-pattern rewrites in the migration CLI

codingby fritzandfriends

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/fritzandfriends-semantic-pattern-guardrails/SKILL.md

decimalai skills pull fritzandfriends-semantic-pattern-guardrails --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/fritzandfriends-semantic-pattern-guardrails/SKILL.md https://app.decimal.ai/s/fritzandfriends-semantic-pattern-guardrails/SKILL.md

Raw file: /s/fritzandfriends-semantic-pattern-guardrails/SKILL.md ↗ · machine-readable summary at /s/fritzandfriends-semantic-pattern-guardrails.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 1h ago

Yes — +50% lift-27% tokens0% turns95% of whole cases pass with skill (21/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	18%	0%
case-02	✗→✓	▲ Improved	-24%	0%
case-04	✗→✓	▲ Improved	-37%	0%
case-08	✗→✓	▲ Improved	-23%	0%
case-10	✗→✓	▲ Improved	-24%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

fritzandfriends / blazorwebformscomponents

.squad/skills/semantic-pattern-guardrails/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

Skill contents

SKILL.md

Context

Use this when adding semantic catalog entries to the Web Forms migration CLI. These rules apply to cross-file page-shape rewrites where the mechanical transforms have already converted raw syntax and the remaining work is contract normalization.

Patterns

Master/content normalization belongs in the semantic pass

Let the markup transforms do the local tag conversion first.
Then normalize generated master/page pairs to the BWFC-tested contract:
master shell markup → ChildContent
migrated <Content ContentPlaceHolderID="..."> blocks → ChildComponents
preserve exact placeholder IDs, including head slots

Query/detail pages must stay read-only

Safe matches:

SelectMethod/code-behind only filters by [QueryString] / [RouteData]
no writes, redirects, form access, auth calls, or postback branches
one clear data-display page shape

Stop and emit TODO/manual review if:

the page mutates state
the page mixes route/query entry points in a way you cannot preserve exactly
the code-behind depends on postback/event sequencing

Action pages must still act

Safe matches:

empty or scaffold-only page
action happens on load from deterministic query inputs
one redirect target
no auth cookies, no POST body dependency, no external provider workflow

Anti-pattern:

converting an action page into a visible informational page that does nothing

Account pages are high-risk and mostly manual

Treat these as manual/TODO boundaries unless you have explicit HTTP endpoint/handler output:

login/logout
register
password reset token issuance/consumption
external auth challenge/callback
2FA provider selection and verification
add/remove external logins
phone / security settings that resignal identity state

Why:

Web Forms account pages depend on live HTTP response semantics and OWIN/Identity middleware
a Blazor component event handler cannot safely fake that contract

Minimum valid output contract

Master shell

razor
<MasterPage>
    <Head>...</Head>
    <ChildContent>
        ...chrome...
        <ContentPlaceHolder ID="MainContent" />
    </ChildContent>
    <ChildComponents>
        @ChildComponents
    </ChildComponents>
</MasterPage>

Child page

razor
<Site>
    <ChildComponents>
        <Content ContentPlaceHolderID="MainContent">
            ...
        </Content>
    </ChildComponents>
</Site>

Anti-Patterns

Rewriting multi-placeholder masters to @Body
Dropping ContentPlaceHolderID names
Pretending auth pages are fully migrated when only the shell was understood
Converting redirect-only action pages into dead-end content pages
Collapsing query/route precedence to a single hard-coded id path

Want to see if fritzandfriends/semantic-pattern-guardrails would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.