First-party app · Ops
ops
An on-call first responder that walks the shift with you. It files each alarm your monitor writes as an incident, ranks the queue by what needs a human first, and works each one with you — claim, investigate against your own runbook, record findings, propose a fix. Proposing changes nothing, and running a fix is off by default, needs an explicit confirm, and can only ever be a command your own runbook already declared.
Before you install
A first-party app of PersonalClaw v0.3.9
No scan on record before install
PersonalClaw publishes no scan of its own apps before you install one.
PersonalClaw runs its own scan of the actual code of every app it installs: this one is rescanned when you install, and that scan is what can stop an install.
Its row on Compare: install rescan
- Compatibility
- Works with PersonalClaw v0.3.9
What it gets
- Permissions
- storage: true
- cron: true
- agent: "tools"
- network: false
- Its own network permission
- Off. Its own network permission is a declaration, not a fence.
What it starts
The programs on this machine it starts, as install consent names them.
The programs you name for it: Applying a fix runs the command a runbook action of yours declares, exactly as your runbook gives it.
While Allow gated remediation is on, each program it starts runs with your own sign-in and settings for that program. Allow gated remediation is off until you turn it on.
What it reaches
- Hosts it names
- Not declared in this release's manifest
Its own code has no host list at this release: any hosts it names are for the programs, servers and sources it starts or fetches, not everything it can reach. Its own network permission is a declaration, not a fence.
Secrets you would enter
- Settings marked secret
- Not declared in this release's manifest
What installing it gives and runs
The manifest's own declarations, under the names install consent shows them by and in its order. The Store adds, at install, which of them run as you.
- Permissions it is granted
{ "storage": true, "cron": true, "agent": "tools", "network": false }- Scheduled jobs
[ { "name": "ops-sweep", "cron_expr": "*/10 * * * *", "message": "Take one on-call sweep.\n\n1. Call ops_watch. If it opened and reopened nothing, stop — say nothing and end the turn.\n2. For each incident it opened or reopened at severity high or critical, or that paged: ops_claim it, then ops_investigate to get its plan.\n3. Walk that plan with your own READ-ONLY tools. Call ops_record once per check with what you actually saw, including the ones that came back clean — a ruled-out cause is worth as much to the next responder as a found one.\n4. When you can name a fix, call ops_propose_fix with the fix, its blast radius and its rollback. If the incident's runbook declares an action that is the fix, name it.\n5. NEVER call ops_apply_fix. An unattended turn does not get to change the system — leave the proposal and its confirm token for a human.\n\nIf you cannot name a fix, record why not and leave the incident claimed. An honest 'not diagnosed, here is what was ruled out' is the useful outcome." } ]- Programs on this machine it starts
[ { "program": "*", "why": "Applying a fix runs the command a runbook action of yours declares, exactly as your runbook gives it.", "inherits": [ "sign-in", "settings" ], "inheritsWhile": { "setting": "allow_apply", "value": true } } ]- Provider modules the gateway loads
[ { "type": "tool", "implementation": "provider:create_provider" } ]- What it runs for personalclaw setup
"app_cli:setup"
- What it runs for personalclaw doctor
"app_cli:doctor"
Not declared in this release's manifest:
- Python packages installed for it
- Engine packages it can install
- What it needs that PersonalClaw does not install
- npm packages installed for it
- Places outside its folder it writes
- Pages it adds to the dashboard
- Components it adds to the dashboard
- A server process of its own
- The sandbox tier its server runs in
- Command run after install
- Command run after an update
- Command run when it is switched on
- Command run when it is switched off
- Command run before it is removed
- Connector sources it parses
- MCP servers it adds
- Skills it installs
Where it comes from
- Source
- https://github.com/PersonalClaw/PersonalClawApps/tree/169cbc4f8fc59cd42a65a6a3546235bc49ab3b22/ops
- Publisher
- PersonalClaw
- License
- MIT
- Pinned commit
- 169cbc4f8fc59cd42a65a6a3546235bc49ab3b22
- Signer
- Not declared in this release's manifest
From review to enable
- Held and scanned. An install stages the app in quarantine, checks its signature, and runs the shared supply-chain scanner over the staged bytes before anything is shown for consent. A registry's recorded verdict is not read at install: the scan that can stop it is this one.Its row on Compare: install rescan
- A dangerous verdict ends it. A dangerous verdict returns a refusal before consent is asked, so there is no consent to give and no switch that skips it. A warning verdict asks for your consent instead.Its row on Compare: dangerous verdict
- You review and consent. The review records a digest of the staged bytes, and an install committed against different bytes is refused: it needs consent again, because the app changed after it was reviewed. An update that changes what the app gets asks for consent again too.Its row on Compare: consent
- Its setup runs, then you switch it on. Nothing it declares runs before your consent; an app already installed is offered as an update instead.
To install it and switch it on
- In PersonalClaw, open the Store.
- Find ops.
- Review what it gets and runs, and consent.
- Install it, then switch it on.
Evidence
ops/app.json at the pinned commit, as written, with the blob hash git lists for it: 1ff713332847fd7708cf7c6152e851c003b812fe
{
"name": "ops",
"version": "0.1.4",
"displayName": "Ops",
"description": "An on-call first responder that walks the shift with you. It files each alarm your monitor writes as an incident, ranks the queue by what needs a human first, and works each one with you — claim, investigate against your own runbook, record findings, propose a fix. Proposing changes nothing, and running a fix is off by default, needs an explicit confirm, and can only ever be a command your own runbook already declared.",
"icon": "Siren",
"author": "PersonalClaw",
"license": "MIT",
"minPersonalClawVersion": "0.2.3",
"tags": [
"tool",
"ops",
"on-call",
"incidents"
],
"cli": {
"setup": "app_cli:setup",
"doctor": "app_cli:doctor"
},
"permissions": {
"storage": true,
"cron": true,
"agent": "tools",
"network": false
},
"launches": [
{
"program": "*",
"why": "Applying a fix runs the command a runbook action of yours declares, exactly as your runbook gives it.",
"inherits": [
"sign-in",
"settings"
],
"inheritsWhile": {
"setting": "allow_apply",
"value": true
}
}
],
"crons": [
{
"name": "ops-sweep",
"cron_expr": "*/10 * * * *",
"message": "Take one on-call sweep.\n\n1. Call ops_watch. If it opened and reopened nothing, stop — say nothing and end the turn.\n2. For each incident it opened or reopened at severity high or critical, or that paged: ops_claim it, then ops_investigate to get its plan.\n3. Walk that plan with your own READ-ONLY tools. Call ops_record once per check with what you actually saw, including the ones that came back clean — a ruled-out cause is worth as much to the next responder as a found one.\n4. When you can name a fix, call ops_propose_fix with the fix, its blast radius and its rollback. If the incident's runbook declares an action that is the fix, name it.\n5. NEVER call ops_apply_fix. An unattended turn does not get to change the system — leave the proposal and its confirm token for a human.\n\nIf you cannot name a fix, record why not and leave the incident claimed. An honest 'not diagnosed, here is what was ruled out' is the useful outcome."
}
],
"provider": {
"id": "tools",
"label": "Work an on-call shift's incidents",
"type": "tool",
"implementation": "provider:create_provider",
"capabilities": [
"ops"
],
"settingsSchema": {
"type": "object",
"properties": {
"spool_dir": {
"type": "string",
"default": "",
"x-meta": {
"label": "Alarm spool folder",
"help": "Where your monitor writes alarms — one JSON file per firing. Empty uses a 'spool' folder in this app's own data dir. Files here are only ever read; this app never deletes or moves them."
}
},
"runbooks_dir": {
"type": "string",
"default": "",
"x-meta": {
"label": "Runbooks folder",
"help": "Your runbooks — one JSON file each, its filename being its name. Empty uses a 'runbooks' folder in this app's own data dir. A runbook holds the checks to walk and, optionally, named remediations with a literal argv."
}
},
"allow_apply": {
"type": "boolean",
"default": false,
"x-meta": {
"label": "Allow gated remediation",
"help": "Off means this app can only ever PROPOSE a fix. On lets ops_apply_fix run a remediation your runbook already declared — and only with an explicit confirm plus the token that digests the exact proposal. Leave it off unless you want that."
}
},
"on_call": {
"type": "string",
"default": "",
"x-meta": {
"label": "On-call name",
"help": "Recorded as the owner when an incident is claimed without naming one. Empty records 'on-call'."
}
},
"timeout_secs": {
"type": "integer",
"default": 60,
"minimum": 5,
"maximum": 900,
"x-meta": {
"label": "Remediation Timeout",
"help": "Maximum seconds a confirmed remediation may run before it is given up on.",
"tags": [
"advanced"
]
}
}
}
}
}
}
- Scan verdict
- No scan on record before install
- Scan time
- No scan is published before install
- Blocking findings
- No scan is published before install
- Scanner version
- No scan is published before install
PersonalClaw publishes no scan of its own apps before you install one.
Every app is scanned again when you install it, whatever its listing says.
An install stages the app in quarantine, checks its signature, and runs the shared supply-chain scanner over the staged bytes before anything is shown for consent. A registry's recorded verdict is not read at install: the scan that can stop it is this one.
apps/app_manager.py at v0.3.9A dangerous scan verdict stops the install, and nothing overrides it.
A dangerous verdict returns a refusal before consent is asked, so there is no consent to give and no switch that skips it. A warning verdict asks for your consent instead.
apps/app_manager.py at v0.3.9Consent covers exactly what you reviewed, and an app that changed asks again.
The review records a digest of the staged bytes, and an install committed against different bytes is refused: it needs consent again, because the app changed after it was reviewed. An update that changes what the app gets asks for consent again too.
apps/app_manager.py at v0.3.9A setting an app marks secret is kept in the credential store and never shown back.
The app's settings file holds only a reference, the value lives in the credential store, and every route that shows settings masks it. The app's own code receives the value when it runs.
apps/secret_fields.py at v0.3.9
PersonalClaw/PersonalClawApps @ 169cbc4f8fc5 · ops/app.json