Add .claude/scripts/log-skill-error.sh — the canonical agent error log helper
(writes errorlog.md in DATE | MACHINE | skill | [type] error format, soft-fails).
Three categories: execution failures (default), user corrections (--correction),
and preventable self-inflicted friction (--friction; cite ref= when it repeats a
documented gotcha). Goal: stop paying tokens twice for the same avoidable mistake.
- CLAUDE.md: make logging mandatory for all skills + corrections + friction.
- skill-creator: new skills must wire in the helper (guidance + checklist).
- Retrofit every skill script's genuine failure branches to call the helper
(b2/bitdefender/mailprotector/packetdial/coord python CLIs; remediation-tool
+ onboard365 bash; vault, rmm-auth, post-bot-alert, agy, grok, 1password,
run-onboarding-diagnostic). Handled conditions + self-tests left alone.
- errorlog.md: broaden header to cover skills + harness + corrections; seed this
session's corrections (INKY, Mail.Send token-audience, omnibox-strictness) and
friction (git-bash /tmp, env-persistence, argv-limit, PowerShell var-case).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Auth selection logic:
- Default: prefer cert when cert_thumbprint_b64url + cert_private_key_pem_b64
are present in the vault entry's credentials block; fall back to client_secret.
- REMEDIATION_AUTH=secret -> force client_secret flow.
- REMEDIATION_AUTH=cert -> force cert flow; error if cert fields missing.
- Logs [INFO] auth=cert/secret to stderr so users see which path was taken.
Cert flow signs an RS256 JWT (header includes x5t) via inline Python (PyJWT
+ cryptography), POSTs client_assertion_type +
client_assertion=<jwt> in place of client_secret. Same scope, same cache, same
error handling (AADSTS7000229 still emits the consent URL).
Single sops -d to a mktemp file feeds both field reads to avoid repeated
~1s decrypt invocations on Windows; trap removes plaintext on exit.
Verified end-to-end against tedards.net for all three modes after wiping
/tmp/remediation-tool/.
1. Variable name collision: VAULT_PATH was used for both the SOPS file
relative path (set by case statement) and the vault root override env
var. Renamed env var override to VAULT_ROOT_ENV to avoid collision.
2. Wrong directory depth: CLAUDETOOLS_ROOT was navigating 3 levels up
from scripts/ landing at .claude/ instead of repo root. Fixed to 4
levels (scripts -> remediation-tool -> skills -> .claude -> repo root).
Also added jq as primary vault_path reader (handles Unix paths on Windows),
with cygpath-converted Python fallback.
Bugs discovered during Mac testing 2026-04-21. Windows worked only because
tokens were served from /tmp cache after first acquisition.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add .claude/scripts/vault.sh wrapper (reads vault_path from identity.json)
- get-token.sh + patch-tenant-admin-manifest.sh read identity.json for vault root
- syncro.md uses wrapper via CLAUDETOOLS_ROOT
- CLAUDE.md + ONBOARDING.md document the pattern and prompt for vault_path on onboarding
- identity.json now includes vault_path (D:/vault on DESKTOP-0O8A1RL)
Howard and Mac need vault_path added to their identity.json after pulling.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace hardcoded D:/vault references with candidate-list pattern
that also checks $HOME/vault, ~/.vault, and respects VAULT_PATH
env var override. Fixes vault.sh lookup failures on Mac and
Howard's machine.
Affected: CLAUDE.md, syncro.md, get-token.sh, patch-tenant-admin-manifest.sh
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Root cause: app-only Graph operations (password reset, Exchange REST) require
directory roles on each SP in the customer tenant, not just admin consent.
RoleManagement.ReadWrite.Directory was missing from all app manifests, making
role assignment impossible without manual portal work that was never being done.
Changes:
- patch-tenant-admin-manifest.sh: adds RoleManagement.ReadWrite.Directory to
Tenant Admin app manifest via Management app, grants home-tenant consent
- onboard-tenant.sh: new script — resolves tenant, acquires Tenant Admin token,
assigns Exchange Administrator to Security Investigator SP and User/Auth
Administrator to User Manager SP; --dry-run supported; idempotent
- get-token.sh: detects AADSTS7000229, emits consent URL + onboard-tenant.sh
reminder instead of silent failure
- gotchas.md: onboarding steps at top, tenant table expanded with role columns,
all known tenants updated including martylryan.com (first fully onboarded)
Verified: martylryan.com fully onboarded, password reset to MLR2026!! succeeded
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Windows Store python3 stub returns exit 49 instead of running Python.
Replace with: py (Windows launcher) for actual Python code, jq for
simple JSON extraction. Reorder fallback loops to try py first.
Add Bash(py:*) to settings.local.json allowlist.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Rewrote get-token.sh: tiered app system (investigator/exchange-op/user-manager/tenant-admin/defender)
- Updated SKILL.md, command, gotchas, checklist, graph-endpoints for new app suite
- Cascades breach check: mailbox clean, inbound phishing received by John, DMARC gap noted
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
