fortify

Name: fortify
Author: Borda

23stars3forks

How to install fortify?

borda/ai-rig/fortify

$npx -y skills add Borda/AI-Rig --skill fortify

Installs into the current project.

Summary

Systematic ablation study runner. After research:run finds improvements, fortify identifies component candidates from git diff + diary, creates isolated git worktrees per ablation (main repo never modified), runs metric+guard in each worktree, ranks component importance, and opti

Files · 1

View on GitHub

SKILL.md

1	<objective>
2
3	Ablation study runner — after `/research:run` finds improvements, fortify identifies which components contributed, generates ablation variants (remove one component at a time), runs each in isolated git worktrees (main repo never modified), ranks component importance, optionally generates reviewer Q&A calibrated to venue.
4
5	NOT for: initial optimization loop (use `/research:run`); methodology validation (use `/research:judge`); paper-vs-code consistency (use `/research:verify`); hypothesis generation (use `research:scientist` directly). Fortify runs ablation studies on completed runs only.
6
7	</objective>
8
9	<constants>
10
11	```yaml
12	MAX_ABLATION_CANDIDATES: 8 (ceiling — scientist produces 3–8; --max-ablations caps further)
13	METRIC_TIMEOUT_MS: 360000 (6 min — same as run SKILL.md)
14	GUARD_TIMEOUT_MS: 360000
15	GIT_OP_TIMEOUT_MS: 15000
16	SANITY_DIVERGENCE_PCT: 2.0 (full-variant vs best_metric mismatch threshold)
17	IMPORTANCE_CLASS_CRITICAL: 50.0 (% of full metric lost)
18	IMPORTANCE_CLASS_SIGNIFICANT: 10.0
19	FORTIFY_DIR_BASE: .experiments
20	STATE_DIR_BASE: .experiments/state
21	METRIC_CMD_SOURCE: state.json .config.metric_cmd (campaign's real metric — no fail-open default; env METRIC_CMD overrides)
22	GUARD_CMD_SOURCE: state.json .config.guard_cmd (campaign's real guard — no fail-open default; env GUARD_CMD overrides)
23	```
24
25	</constants>
26
27	Environment overrides — set these before invoking the skill to override per-variant defaults:
28
29	- `METRIC_CMD` — command run inside each variant worktree to measure the ablation metric (override; default sourced from source-run `state.json` `.config.metric_cmd`)
30	- `GUARD_CMD` — command run inside each variant worktree to detect regressions (override; default sourced from source-run `state.json` `.config.guard_cmd`)
31	- `STATE_DIR_BASE` — base directory for source-run state lookups (default: `.experiments/state`)
32	- `FORTIFY_DIR_BASE` — base directory for per-run fortify artifacts (default: `.experiments`)
33
34	The constants block defaults above are YAML-only — bash blocks read environment variables (with `${VAR:-default}` fallback) and never source values directly from YAML.
35
36	<compaction>
37
38	Key boundaries: end of F2 — ablation candidates identified by scientist; end of F4 — all ablation variant worktrees completed.
39	Preserve at F2: FORTIFY_DIR (TMPDIR key), RUN_ID (TMPDIR key), ablation-candidates.jsonl path, best_metric from source run.
40	Preserve at F4: FORTIFY_DIR, RUN_ID, results.jsonl path — ready for F5 importance ranking.
41	F4 loop is resume-safe: 4a-guard skips variants already terminal (non-timeout) in results.jsonl, so a mid-loop compaction resumes at the first pending variant — never re-runs a completed ablation; a timed-out variant still retries.
42	Clear at F1 start (stale prior run) and at start of F8 terminal summary.
43
44	</compaction>
45
46	<workflow>
47
48	<!-- Agent resolution: see _RESEARCH_SHARED/agent-resolution.md -->
49
50	## Agent Resolution
51
52	Agent resolution: load and follow the protocol below. Contains foundry check + fallback table. If foundry not installed: use table to substitute each `foundry:X` with `general-purpose`. Agent this skill dispatches: `research:scientist` (same plugin — no fallback if research plugin installed).
53	```bash
54	# loads: compaction-contract.md
55	_RESEARCH_SHARED=$(python "${CLAUDE_PLUGIN_ROOT:-plugins/cc_research}/bin/resolve_shared.py" 2>/dev/null) # timeout: 5000
56	[ -z "$_RESEARCH_SHARED" ] && { echo "! Plugin path resolution failed — ensure research plugin installed and CLAUDE_PLUGIN_ROOT set, or invoke /research:fortify from project root."; exit 1; }
57	cat "$_RESEARCH_SHARED/agent-resolution.md"
58	```
59
60	\| Agent \| Fallback if absent \|
61	\| --- \| --- \|
62	\| `research:scientist` \| `general-purpose` (ablation candidate identification and reviewer Q&A quality reduced — ⚠ general-purpose agent may not emit the JSON envelope this skill parses; surface partial output and surface ⚠ in F7 report) \|
63
64	## CRITICAL: Worktree-based isolation
65
66	Do NOT use `git checkout -b <branch>` for ablations — dirties main working tree, corrupts concurrent tool calls. Each ablation gets own git worktree under `$FORTIFY_DIR/worktrees/<variant>`, created from `best_commit`. Main working tree NEVER modified. Cleanup: `git worktree remove --force` per variant; `git worktree prune` on interrupt.
67
68	## F

People who liked this also viewed

6 picks

Type

fortify

23stars3forks

View on GitHub

How to install fortify?

borda/ai-rig/fortify

$npx -y skills add Borda/AI-Rig --skill fortify

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	<objective>
2
3	Ablation study runner — after `/research:run` finds improvements, fortify identifies which components contributed, generates ablation variants (remove one component at a time), runs each in isolated git worktrees (main repo never modified), ranks component importance, optionally generates reviewer Q&A calibrated to venue.
4
5	NOT for: initial optimization loop (use `/research:run`); methodology validation (use `/research:judge`); paper-vs-code consistency (use `/research:verify`); hypothesis generation (use `research:scientist` directly). Fortify runs ablation studies on completed runs only.
6
7	</objective>
8
9	<constants>
10
11	```yaml
12	MAX_ABLATION_CANDIDATES: 8 (ceiling — scientist produces 3–8; --max-ablations caps further)
13	METRIC_TIMEOUT_MS: 360000 (6 min — same as run SKILL.md)
14	GUARD_TIMEOUT_MS: 360000
15	GIT_OP_TIMEOUT_MS: 15000
16	SANITY_DIVERGENCE_PCT: 2.0 (full-variant vs best_metric mismatch threshold)
17	IMPORTANCE_CLASS_CRITICAL: 50.0 (% of full metric lost)
18	IMPORTANCE_CLASS_SIGNIFICANT: 10.0
19	FORTIFY_DIR_BASE: .experiments
20	STATE_DIR_BASE: .experiments/state
21	METRIC_CMD_SOURCE: state.json .config.metric_cmd (campaign's real metric — no fail-open default; env METRIC_CMD overrides)
22	GUARD_CMD_SOURCE: state.json .config.guard_cmd (campaign's real guard — no fail-open default; env GUARD_CMD overrides)
23	```
24
25	</constants>
26
27	Environment overrides — set these before invoking the skill to override per-variant defaults:
28
29	- `METRIC_CMD` — command run inside each variant worktree to measure the ablation metric (override; default sourced from source-run `state.json` `.config.metric_cmd`)
30	- `GUARD_CMD` — command run inside each variant worktree to detect regressions (override; default sourced from source-run `state.json` `.config.guard_cmd`)
31	- `STATE_DIR_BASE` — base directory for source-run state lookups (default: `.experiments/state`)
32	- `FORTIFY_DIR_BASE` — base directory for per-run fortify artifacts (default: `.experiments`)
33
34	The constants block defaults above are YAML-only — bash blocks read environment variables (with `${VAR:-default}` fallback) and never source values directly from YAML.
35
36	<compaction>
37
38	Key boundaries: end of F2 — ablation candidates identified by scientist; end of F4 — all ablation variant worktrees completed.
39	Preserve at F2: FORTIFY_DIR (TMPDIR key), RUN_ID (TMPDIR key), ablation-candidates.jsonl path, best_metric from source run.
40	Preserve at F4: FORTIFY_DIR, RUN_ID, results.jsonl path — ready for F5 importance ranking.
41	F4 loop is resume-safe: 4a-guard skips variants already terminal (non-timeout) in results.jsonl, so a mid-loop compaction resumes at the first pending variant — never re-runs a completed ablation; a timed-out variant still retries.
42	Clear at F1 start (stale prior run) and at start of F8 terminal summary.
43
44	</compaction>
45
46	<workflow>
47
48	<!-- Agent resolution: see _RESEARCH_SHARED/agent-resolution.md -->
49
50	## Agent Resolution
51
52	Agent resolution: load and follow the protocol below. Contains foundry check + fallback table. If foundry not installed: use table to substitute each `foundry:X` with `general-purpose`. Agent this skill dispatches: `research:scientist` (same plugin — no fallback if research plugin installed).
53	```bash
54	# loads: compaction-contract.md
55	_RESEARCH_SHARED=$(python "${CLAUDE_PLUGIN_ROOT:-plugins/cc_research}/bin/resolve_shared.py" 2>/dev/null) # timeout: 5000
56	[ -z "$_RESEARCH_SHARED" ] && { echo "! Plugin path resolution failed — ensure research plugin installed and CLAUDE_PLUGIN_ROOT set, or invoke /research:fortify from project root."; exit 1; }
57	cat "$_RESEARCH_SHARED/agent-resolution.md"
58	```
59
60	\| Agent \| Fallback if absent \|
61	\| --- \| --- \|
62	\| `research:scientist` \| `general-purpose` (ablation candidate identification and reviewer Q&A quality reduced — ⚠ general-purpose agent may not emit the JSON envelope this skill parses; surface partial output and surface ⚠ in F7 report) \|
63
64	## CRITICAL: Worktree-based isolation
65
66	Do NOT use `git checkout -b <branch>` for ablations — dirties main working tree, corrupts concurrent tool calls. Each ablation gets own git worktree under `$FORTIFY_DIR/worktrees/<variant>`, created from `best_commit`. Main working tree NEVER modified. Cleanup: `git worktree remove --force` per variant; `git worktree prune` on interrupt.
67
68	## F

People who liked this also viewed

6 picks

Type