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 commands/majiayu000/vibeguard/exec-plangit clone --depth 1 https://github.com/majiayu000/vibeguardWrote 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/commands/majiayu000/vibeguard/exec-plan)<a href="https://agentmods.dev/commands/majiayu000/vibeguard/exec-plan"><img src="https://agentmods.dev/badge/commands/majiayu000/vibeguard/exec-plan.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 | $0.00025 | $0.01964 |
| Opus 5 | $0.00013 | $0.00982 |
| Sonnet 5 | $0.00005 | $0.00393 |
| Haiku 4.5 | $0.00003 | $0.00196 |
Grade A, and why
VibeGuard: ExecPlan scanned grade A with 1 finding 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 3d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Nyquist Rule**: Each Step must contain the `verify_cmd` field - a verification command that can be executed within 60 seconds (such as `cargo test --lib`, `curl localhost:8080/health`). Steps that cannot be verified How it starts
The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Core Concept (from OpenAI Harness Engineering)
- Long-term tasks require self-contained execution documents that can be resumed in new sessions by themselves
- Progress is the only chapter that allows checklist, the rest are described in prose
- Decision Log records all decisions that deviate from SPEC to ensure traceability
- ExecPlan is a living document that is continuously updated with execution
Three modes
| Mode | Usage | Description |
|---|---|---|
init |
/vibeguard:exec-plan init [spec path] |
Generate ExecPlan from SPEC |
update |
/vibeguard:exec-plan update <execplan path> |
Append Discovery/Decision/completion status |
status |
/vibeguard:exec-plan status <execplan path> |
View Progress progress summary |
Trigger condition
- SPEC generated and confirmed via
/vibeguard:interview - Tasks expected to be completed across 2+ sessions
- Scenarios where execution context needs to be restored across sessions
Guardrails
initmode does not make any code modifications, only generates documentationupdatemode only modifies the ExecPlan file itself- Does not replace preflight - ExecPlan defines "what to do", preflight defines "what not to do"
Mode: init
Generate ExecPlan files from SPEC.
Steps
-
Read SPEC
- If spec path ($ARGUMENTS) is provided, read the file
- If not provided, search for
SPEC.mdin the project root directory - If there is no SPEC, prompt the user to run
/vibeguard:interviewfirst
-
Analyze SPEC and break down milestones
- Extract milestones from SPEC functional requirements (FR-XX)
- Each milestone contains 1-3 specific steps
- Identify dependencies between milestones
-
Scan project context
- Identify languages/frameworks and key entry files
- Check if there is a preflight constraint set to reference
- Document existing code locations related to SPEC
3.5. Capture the W-20 execution snapshot
- Create
${VIBEGUARD_HOME:-${HOME}/.vibeguard}/artifacts/execplan/<project-name>/outside the target repository, then write the active runtime, model, tool/MCP/skill names, and stable description hashes to its<task>-tool-inventory.txt; each non-comment line must be<kind> <name> <description_sha256> - Resolve the installed source as
${VIBEGUARD_DIR:-${HOME}/.vibeguard/installed}and verify that its drift guard andrules/claude-rulesdirectory exist - Run
bash "${VIBEGUARD_DIR:-${HOME}/.vibeguard/installed}/guards/universal/check_runtime_drift.sh" snapshot --snapshot "${VIBEGUARD_HOME:-${HOME}/.vibeguard}/artifacts/execplan/<project-name>/<task>-runtime.snapshot" --tool-inventory "${VIBEGUARD_HOME:-${HOME}/.vibeguard}/artifacts/execplan/<project-name>/<task>-tool-inventory.txt" --rules-dir "${VIBEGUARD_DIR:-${HOME}/.vibeguard/installed}/rules/claude-rules" - Stop if the inventory or snapshot cannot be created; do not generate a resumable ExecPlan without them
-
Generate ExecPlan
- Populate 8 chapters by template (
workflows/plan-flow/references/execplan-template.md) - Purpose extracted directly from the SPEC overview
- Progress is mapped to a milestone list with checkbox
- Concrete Steps aligns the Step format of plan-template.md (status/target/file/change/test/judgment)
- Nyquist Rule: Each Step must contain the
verify_cmdfield - a verification command that can be executed within 60 seconds (such ascargo test --lib,curl localhost:8080/health). Steps that cannot be verified within 60s are marked asunverifiableand need to be split or supplemented with verification methods. - Validation converted from SPEC acceptance criteria (AC-XX)
- Decision Log is initially empty and records the selection decisions during generation.
- Record the W-20 snapshot and tool-inventory paths in Context.
- Populate 8 chapters by template (
-
Save and Confirm
- Save to
<project name>-execplan.md(project root directory) - Display Progress and Concrete Steps summaries for user confirmation
- Use AskUserQuestion to confirm if adjustments are needed
- Save to
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.
- 3d ago First seen · 156 lines · 25 tokens per session scan A 50c4059cf170
VibeGuard: ExecPlan is a command published in the GitHub repository majiayu000/vibeguard (41 stars, last pushed 4d ago), licensed MIT. It adds 25 tokens to every session and 1,964 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
patterns
List makoto's integrity pattern catalog (read-only).
show
Read makoto's results ledger by key (read-only).
status
Show makoto's integrity-guardrail state (read-only).
claudemd-design-adopt
Generate a thin, fact-based DESIGN.md from a UI project's real design-token sources and wire it into project CLAUDE.md. Use when (1) the user asks to configure design specs / 配置设计规范 / 设计规范 for agents, (2) UI sessions keep inventing colors or spacing instead of using the project's existing tokens, (3) a UI project…
claudemd-refresh
Refresh the installed claudemd plugin to the latest released version in one shot (marketplace update → uninstall → install via the claude CLI). Use when the SessionStart banner reports a newer version, or /claudemd-doctor flags a stale plugin cache. Restart Claude Code afterwards — spec + manifest then sync…
claudemd-update
Sync /.claude/CLAUDE.md with the plugin-cache shipped spec. Shows diff summary; user chooses apply-all or cancel.