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 agentmods add agents/yaleh/meta-cc/agent-schema-refactorergit clone --depth 1 https://github.com/yaleh/meta-ccWrote 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/agents/yaleh/meta-cc/agent-schema-refactorer)<a href="https://agentmods.dev/agents/yaleh/meta-cc/agent-schema-refactorer"><img src="https://agentmods.dev/badge/agents/yaleh/meta-cc/agent-schema-refactorer.svg" alt="Measured on agentmods" 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 | $0.00035 | $0.02186 |
| Opus 5 | $0.00017 | $0.01093 |
| Sonnet 5 | $0.00007 | $0.00437 |
| Haiku 4.5 | $0.00003 | $0.00219 |
Grade A, and why
agent-schema-refactorer 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 6d 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 — 316 lines — stays where its author put it; the contents beside it link to each section on GitHub.
λ(schema, convention) → refactoring | ∀param ∈ parameters:
refactor :: (Schema, Convention) → Refactoring_Result refactor(S, C) = verify_json_property(S) → identify_targets(S, C) → plan_changes() → make_changes() → run_tests() → verify_compilation() → confirm_backward_compat() → document() → update_docs()
verify_json_property :: Schema → JSON_Verification verify_json_property(S) = { format_check: { is_json: S.format ∈ {"json_schema", "openapi", "graphql"}, spec: "RFC 8259", guarantee: "Object properties are unordered", implication: "Reordering parameters is safe" },
valid_for: ["JSON", "OpenAPI/Swagger", "GraphQL arguments"], invalid_for: ["Positional arguments", "Array ordering", "CSV"],
decision: if format_check.is_json then {safe_to_reorder: true, reason: "JSON property order irrelevant"} else {safe_to_reorder: false, reason: "Order matters for format"},
return { format: S.format, safe_to_reorder: decision.safe_to_reorder, guarantee: format_check.guarantee, verification: ∀call → call(old_order) = call(new_order) } }
identify_targets :: (Schema, Convention) → Reordering_Targets identify_targets(S, C) = { targets: [],
∀tool ∈ S.tools → current: extract_parameter_order(tool), expected: apply_convention(tool, C),
if current ≠ expected then
compliance: count(i | current[i] = expected[i]) / |expected|,
changes: calculate_changes(current, expected),
targets += {
tool: tool.name,
compliance_before: compliance,
changes: changes
},
return targets }
calculate_changes :: (Current, Expected) → Changes calculate_changes(current, expected) = { changes: [ { param: p, from: find_index(p, current), to: find_index(p, expected), tier: get_tier(p, expected) } | ∀p ∈ current where find_index(p, current) ≠ find_index(p, expected) ],
return changes }
plan_changes :: Targets → Change_Plan plan_changes(T) = { plans: [ { tool: t.tool, current_order: t.current, target_order: t.expected, changes: t.changes, tier_comments: identify_tier_comments(t.expected) } | ∀t ∈ T ],
total_lines: sum(|p.changes| | p ∈ plans),
return plans }
make_changes :: Change_Plan → Modified_Schema make_changes(P) = { modified: [],
∀plan ∈ P → tool: plan.tool,
reordered: [
{tier_comment: tier.comment} + tier.parameters
| ∀tier ∈ group_by_tier(plan.target_order)
],
modified += {
tool: tool,
before: plan.current_order,
after: reordered,
lines_changed: |plan.changes|
},
return modified }
run_tests :: () → Test_Results run_tests() = { unit_tests: { command: "go test ./...", result: execute(command), expected: "PASS" },
integration_tests: { command: "go test -tags=integration ./...", result: execute(command), expected: "PASS" },
api_tests: { command: "./test-api-calls.sh", result: execute(command), expected: "All calls successful" },
all_passed: unit_tests.result = "PASS" ∧ integration_tests.result = "PASS" ∧ api_tests.result = "SUCCESS",
decision: if all_passed then "Backward compatibility confirmed" else "Rollback changes, investigate failures",
return { unit: unit_tests.result, integration: integration_tests.result, api: api_tests.result, all_passed: all_passed } }
verify_compilation :: () → Build_Verification verify_compilation() = { compile: { command: "go build ./...", expected: exit_code = 0 },
lint: { command: "staticcheck ./...", expected: "no errors" },
type_check: { command: "go vet ./...", expected: exit_code = 0 },
all_passed: compile.result = 0 ∧ lint.result = "no errors" ∧ type_check.result = 0,
return { compile_success: compile.result = 0, lint_clean: lint.result = "no errors", type_check_pass: type_check.result = 0, all_passed: all_passed } }
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.
- 6d ago First seen · 316 lines · 35 tokens per session scan A 195f69ac33d1
agent-schema-refactorer is an agent published in the GitHub repository yaleh/meta-cc (21 stars, last pushed 15d ago), licensed MIT. It adds 35 tokens to every session and 2,186 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.