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 skills/gowtham012/claude-plugins/solvenpx skills add gowtham012/Claude-plugins --skill solvegit clone --depth 1 https://github.com/gowtham012/Claude-pluginsWhat 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 | $0.00033 | $0.02141 |
| Opus 5 | $0.00016 | $0.01071 |
| Sonnet 5 | $0.00007 | $0.00428 |
| Haiku 4.5 | $0.00003 | $0.00214 |
Grade A, and why
solve 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 2d 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 — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Problem from user
$ARGUMENTS
Current project context (auto-injected)
Language/framework detection:
!`ls package.json pyproject.toml Cargo.toml go.mod pom.xml Gemfile composer.json mix.exs 2>/dev/null || echo "(no package manifest found)"`
Existing test directory:
!`ls -d tests/ test/ __tests__/ spec/ src/test/ src/__tests__/ 2>/dev/null | head -10 || echo "(no test directory found)"`
Active task (if resuming):
!`cat deep-engineer/current-task.md 2>/dev/null || echo "(no active task)"`
Discipline warnings (if any):
!`grep -c '"warning"' deep-engineer/log.jsonl 2>/dev/null || echo "0"` warnings in log
RESUME LOGIC — Check this FIRST
Read the "Active task" section above.
If there IS an active task, determine the current phase from the frontmatter (phase: <value>) and skip directly to that phase's section below. Do NOT repeat completed phases. Announce:
Resuming task at Phase N:
Previously completed:
If there is NO active task, start from Phase 1.
CRITICAL RULES — Read before doing ANYTHING
- The user's description is ONE example. Identify the GENERAL pattern.
- You MUST list at least 5 scenarios before writing any code. The
save_tasktool will REJECT fewer than 5. - You MUST NOT write implementation code before tests. The stop hook will BLOCK you and print a violation.
- Your tests MUST include cases the user did NOT mention.
- If your code uses any value from the user's specific example as a literal/hardcode, you're doing it wrong.
- You MUST run tests and loop until ALL pass.
- You CANNOT skip phases. The
update_phasetool enforces sequential order. - Check for warnings by calling
check_warningsbefore each phase transition.
Phase 1: Generalize
Restate the user's problem as a GENERAL problem. Strip away their specific example and find the underlying pattern.
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.
- 2d ago First seen · 238 lines · 33 tokens per session scan A f269d9cd2745
solve is a skill published in the GitHub repository gowtham012/Claude-plugins (5 stars, last pushed 5mo ago), licensed MIT. It adds 33 tokens to every session and 2,141 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 2026-08-31.
Other skills, from other repositories
specify-incremental
Decompose a single-feature specification into a linear, phase-by-phase implementation plan. Use this for medium-complexity work — single feature, one or two components — where transparent human-in-the-loop phase review is preferred over factory automation.
execute
Dispatch and execute implementation plans with TDD and checkpoints. Use when plan is ready. Parallel by default for independent tasks.
tdd
Strict test-driven development for behavior changes. Requires verified RED before production code, minimal GREEN, and refactor only after passing tests.
doncheli-implement
Execute TDD implementation from task list following RED-GREEN-REFACTOR cycle with mandatory tests, stub detection, and quality gates. Activate when user mentions "implement", "build", "code", "develop", "implementar".
xp-tdd-practices
Read this guideline when creating implementation tasks, doing TDD, deciding task order, or working on any feature with observable behavior.
pdlc-feature
全自动 PDLC 新功能开发（串联 PRD→设计→TDD→实现→评审→发布）.