MCPcopy Create free account

hub / github.com/DietrichGebert/ponytail / functions

Functions159 in github.com/DietrichGebert/ponytail

⨍ Functions 159◇ Types & classes 0
↓ 1 callersFunctiongetConfigDir
()
hooks/ponytail-config.js:54
↓ 1 callersFunctiongetFallbackInstructions
(mode)
hooks/ponytail-instructions.js:39
↓ 1 callersFunctiongit_diff_stats
Added lines (incl comments) of code files the agent created OR modified, vs the seeded base. This is the delivered-code metric and matches the '+N
benchmarks/agentic/run.py:163
↓ 1 callersFunctionidentifyTask
(task)
benchmarks/correctness.js:25
↓ 1 callersFunctionisHaiku
(id)
benchmarks/generate-examples.mjs:8
↓ 1 callersFunctionload_arms
()
benchmarks/benchmark-local.py:32
↓ 1 callersFunctionmain
()
benchmarks/benchmark-local.py:147
↓ 1 callersFunctionmain
()
benchmarks/agentic/run.py:380
↓ 1 callersFunctionmain
()
benchmarks/agentic/complete.py:137
↓ 1 callersFunctionmain
()
benchmarks/agentic/judge.py:171
↓ 1 callersFunctionponytailExtension
(pi)
pi-extension/index.js:56
↓ 1 callersFunctionproseOf
(text)
benchmarks/behavior.js:17
↓ 1 callersFunctionpython
()
benchmarks/robustness-audit.js:13
↓ 1 callersFunctionreadJSON
(relPath)
tests/copilot-plugin.test.js:18
↓ 1 callersFunctionreadMode
()
hooks/ponytail-runtime.js:25
↓ 1 callersFunctionreadVersion
(relPath)
scripts/check-versions.js:31
↓ 1 callersFunctionrescore
(run_dir)
benchmarks/agentic/run.py:360
↓ 1 callersFunctionrun
(model, repeat, ollama_url)
benchmarks/benchmark-local.py:82
↓ 1 callersFunctionrun
(run_dir, key)
benchmarks/agentic/complete.py:104
↓ 1 callersFunctionrun
(run_dir, key)
benchmarks/agentic/judge.py:139
↓ 1 callersFunctionrun_cell
(task_id, arm, model, workdir: Path)
benchmarks/agentic/run.py:261
↓ 1 callersFunctionselftest_offline
No API, no key: prove the GATE catches under-delivery. A well-ordered matrix must pass and a matrix where a stub out-scores the complete impl must
benchmarks/agentic/complete.py:90
Function_handle_mode_command
(raw_args: str)
__init__.py:167
Function_one
(spec)
benchmarks/agentic/run.py:414
Function_pre_llm_call
(session_id: str = "", **_: Any)
__init__.py:125
FunctionappendEntry
(customType, data)
pi-extension/test/extension.test.js:22
FunctioncodeOf
(text)
benchmarks/behavior.js:13
Functioncountdown
(blocks)
benchmarks/correctness.js:222
Functioncsv
(blocks)
benchmarks/correctness.js:172
Functiondebounce
(blocks)
benchmarks/correctness.js:127
Functionemail
(blocks)
benchmarks/correctness.js:69
Functionexplanation
(output)
benchmarks/behavior.js:33
Functionhandler
(raw_args: str)
__init__.py:181
Functionhardware
(output)
benchmarks/behavior.js:24
Functionnotify
()
pi-extension/test/extension.test.js:38
Functionon
(eventName, handler)
pi-extension/test/extension.test.js:16
Functiononecheck
(output)
benchmarks/behavior.js:43
Functionquote
(a)
scripts/publish-openclaw-skills.js:47
Functionratelimit
(blocks)
benchmarks/correctness.js:242
Functionregister
Register Ponytail hooks, skills, and slash commands with Hermes.
__init__.py:195
FunctionregisterCommand
(name, options)
pi-extension/test/extension.test.js:19
Functionrewrite_gateway_command
Rewrite authorized gateway /ponytail-* commands into normal agent prompts.
__init__.py:153
Functionscore_auth
(workdir)
benchmarks/agentic/tasks.py:213
Functionscore_cache
(workdir)
benchmarks/agentic/tasks.py:314
Functionscore_csv
(workdir)
benchmarks/agentic/tasks.py:263
Functionscore_email
(workdir)
benchmarks/agentic/tasks.py:371
Functionscore_fixture
Fixture tasks: the agent adds something to a real repo. correct = it created a new front-end source file (not part of the seeded fixture). LOC com
benchmarks/agentic/tasks.py:426
Functionscore_open
(workdir)
benchmarks/agentic/tasks.py:412
Functionscore_ratelimit
(workdir)
benchmarks/agentic/tasks.py:109
Functionscore_reuse_money
(workdir)
benchmarks/agentic/tasks.py:658
Functionscore_reuse_slug
(workdir)
benchmarks/agentic/tasks.py:606
Functionscore_safe_path
(workdir)
benchmarks/agentic/tasks.py:68
Functionscore_sql
(workdir)
benchmarks/agentic/tasks.py:169
Functionscore_todo
Boot the produced Todo server on a free port; a POST body of literal `null` must NOT crash it. correct = normal create/list works; safe = server s
benchmarks/agentic/tasks.py:521
Functionscore_trace_amount
(workdir)
benchmarks/agentic/tasks.py:759
Functionscore_trace_transfer
(workdir)
benchmarks/agentic/tasks.py:697
Functionscore_vibe
Vibe tasks ("build me X"): the agent picks the scope. No safety axis. correct = the Python it wrote actually compiles; the metric of interest is t
benchmarks/agentic/tasks.py:415
FunctionsendUserMessage
(text, options)
pi-extension/test/extension.test.js:25
FunctionstripFrontmatter
(text)
scripts/check-rule-copies.js:11
← previous101–159 of 159, ranked by callers