AXIOM — Autonomous Political Engagement System

Section 01

What is AXIOM?

AXIOM is an autonomous political engagement system built to amplify authentic values-driven commentary on Australian politics. It operates as an always-on AI agent running on a private server, managing a Twitter/X account with three coordinated activity layers — original content generation, targeted intellectual engagement, and broad topic-level visibility.

Unlike simple scheduling tools, AXIOM treats every interaction as a data point in a longitudinal study. It is simultaneously a communications platform and a research instrument — tracking exactly what messaging style, emotional register, topic framing, and timing drives engagement with specific audience segments.

🗞️ Content Layer

Scans AU political news feeds in real time. Generates opinionated, voice-consistent commentary. Posts 2–6 times daily at irregular intervals within peak engagement windows.

🧠 Thinker Layer

Monitors 10 curated AU left intellectuals. Identifies their most substantive recent tweets. Generates intellectually engaged replies — extending arguments, asking productive questions, adding the behavioural economics layer.

🔍 Scanner Layer

Searches hashtags and keywords across 8 topic domains. Finds posts gaining traction from non-curated accounts. Comments to build visibility with audiences who don't yet know the account.

Section 02

Daily Operating Schedule

All times in Sydney (AEST/AEDT). The system runs autonomously — no human intervention required.

06:00 — 01:00 (every 20 min)

Content Pipeline + Posting Scheduler

Fetches AU political news, classifies by topic, generates opinionated commentary in the account voice, queues and posts. Probabilistic timing model — fires within peak windows (morning commute 7–9am, lunch 12–1pm, evening 5–7pm, prime time 9–11pm) with randomised jitter to avoid detectable bot patterns.

08:30 / 12:30 / 20:30

Topic Scanner

Searches #auspol, #climate, #ndis, #housing and 5 other topic streams. Scores posts for engagement worthiness (sweet spot: 20–300 likes). Comments on up to 5 posts per day across all runs. Avoids viral noise and obvious bait.

09:00 / 17:00 / 21:00

Thinker Engagement

Checks recent tweets from Richard Denniss, Greg Jericho, Ketan Joshi, Antony Loewenstein, Lizzie O'Shea, Teela Reid, Adam Triggs, Bill Hare, Michael West, Margaret Simons. Generates substantive replies (extend / question / bridge / challenge modes). Max 3 thinker engagements per day.

Every 2 hours

Token Refresh

Automatically refreshes the Twitter OAuth2 access token. Zero manual intervention ever required.

Every 30 min (06:00–00:00)

Engagement Tracker

Polls Twitter API for new mentions. Updates engagement metrics (likes, reposts, replies, impressions) on all posted tweets. Records interacting users in the audience panel.

23:00

Daily Metrics Snapshot

Takes a timestamped snapshot of all aggregate metrics. Builds the 30-day trend history used for longitudinal analysis.

Section 03

What the System Captures

3.1 — Per-Tweet Classification (v3 Schema)

Every tweet posted by the system is logged with the following dimensions — all available for later correlation against engagement outcomes.

Dimension	Values	Source
Emotional Tone	earnestwrypassionatesorrowfulsardonichopefulmeasuredprovocative	LLM-classified
Sentiment	criticalsupportivequestioninghopefulangrycelebratoryneutralambivalent	LLM-classified
Style Register	casualanalyticalliterarypunchyconversational	Computed
Sophistication Level	1 (accessible) → 4 (specialist)	Computed from word/sentence complexity
Emotion/Reason Ratio	0.0 (pure reason) → 1.0 (pure emotion)	Lexicon-computed
Rhetorical Device	rhetorical_questionironystatistic_anchorreframecall_to_actionparadox	Computed
Content Type	original_opinion / news_reaction / thinker_reply / topic_scan_reply / quote_tweet	System-assigned
Posting Context	breaking_news / ongoing_issue / policy_debate / unprompted	System-assigned
Timing	Hour of day (0–23), Day of week (Mon–Sun)	System timestamp
Topic (Hierarchical)	Domain → Category → Subcategory (see Section 3.2)	Keyword-classified
Message Metrics	Word count, char count, sentence count, avg word length, avg sentence length, hashtag count, has_question, has_link	Computed

3.2 — Hierarchical Topic Taxonomy

Topics are classified at three levels. Example paths shown below.

EXAMPLE PATHS

        Foreign Policy → Middle East → Palestine / Gaza

        Foreign Policy → Middle East → Israel

        Economy → Housing → Negative Gearing

        Economy → Work and Wages → Minimum Wage

        Social Policy → Disability → NDIS

        War and Peace → Active Conflicts → Gaza / Palestine

        Indigenous Affairs → Sovereignty → Treaty

        Environment → Fossil Fuels → Greenwash

DOMAINS (9 TOP-LEVEL)

Domestic Politics Economy Social Policy Environment & Climate Indigenous Affairs Foreign Policy War and Peace Human Rights Media & Democracy Corporate Power

3.3 — Engagement Metrics (Time-Series)

Every 30 minutes, the engagement tracker polls the API and appends a snapshot to each tweet's record. This gives a time-series of engagement growth — not just a point-in-time count.

Metric	Description	Update Frequency
Likes	Cumulative likes on the tweet	Every 30 min
Reposts	Retweets (weighted 2x in engagement score)	Every 30 min
Replies	Replies received	Every 30 min
Quotes	Quote tweets from other accounts	Every 30 min
Impressions	Total views (when available via API)	Every 30 min
Composite Score	likes + (reposts × 2) + replies	Computed on query

3.4 — Audience Panel (Longitudinal)

Every user who interacts with the account is entered into a longitudinal panel. Returning users receive a new observation appended to their record — building a true panel dataset that grows in both size and depth over time.

Privacy architecture: Individual records are never surfaced. Only aggregates are queryable. EU/EEA users are excluded at ingestion (GDPR compliance). Raw Twitter user IDs are replaced with a one-way SHA256 hash that cannot be reversed. All inferred fields are tagged is_imputed: true with confidence scores and basis strings.

Field	Source	Imputed?	Confidence
User Hash	SHA256 of Twitter ID	No	—
Follower Count	Twitter public API	No	100%
Account Age	Twitter created_at	No	100%
Country	Location string → keyword match	Sometimes	45–85%
AU State / City	Location string → keyword match	Sometimes	60–85%
Gender	First name database + bio pronouns	Yes	70–90%
Age Range	Bio signals + account age heuristics	Yes	40–65%
Political Lean	Bio keyword signals	Yes	50–88%
Profession	Bio keyword signals	Yes	20–62%
Issue Interests	Bio keyword signals	Yes	20–68%
Interaction Count	Panel observations	No	100%

Section 04

Sample Analytics Reports

The following illustrate the kind of analysis available once the panel accumulates data. Numbers shown are illustrative projections based on typical AU political account growth trajectories.

4.1 — Week 4 Snapshot (Projected)

112

Tweets Posted

847

Total Likes

234

Reposts

340

New Followers

189

Panel Members

31%

Repeat Interactors

7.6

Avg Likes / Tweet

22%

Thinker Reply Rate

4.2 — Engagement by Emotional Tone

Which tone drives the most engagement? Composite score = likes + (reposts × 2) + replies.

Wry

14.2 avg

Passionate

11.8 avg

Hopeful

10.9 avg

Earnest

9.8 avg

Measured

7.2 avg

Sorrowful

5.7 avg

Sardonic

4.1 avg

4.3 — Engagement Over Time (30-Day Trend)

Daily composite engagement score (likes + reposts×2 + replies) and follower growth over the first 30 days. Projected trajectory based on typical AU political account benchmarks.

4.3b — Sentiment Distribution Over Time

How the emotional register of posts shifts week by week. The system varies tone — tracking which sentiment mix correlates with peak engagement periods.

4.3c — Topic Volume Over Time

How much content is posted per domain each week — driven by news cycles. War & Peace spikes when conflict escalates; Economy spikes around budget season.

4.3d — Emotion/Reason Ratio vs Engagement

Each point is one tweet. X-axis = emotion ratio (0 = pure reason, 1 = pure emotion). Y-axis = composite engagement score. Reveals the optimal emotional register for this audience.

4.3e — Issue Tracking: Audience Interest vs Content Volume

Radar chart comparing what topics the audience cares about (inferred from bio signals) vs what we're posting about. Gap analysis — where are we under- or over-serving audience interests?

4.3f — Network Diffusion: How Ideas Spread

Tracks propagation topology — how content moves through the social graph. Nodes are classified by structural position (influence tier, network role) without individual identification. All actor data is irreversibly hashed.

Amplifier Influence Tiers

Who is spreading our content? Classified by follower count. Macro = 100k+, Mid = 10k–100k, Micro = 1k–10k, Nano = 100–1k.

Propagation Depth Distribution

How many hops does content travel? Hop 0 = our original post. Hop 1 = direct repost. Hop 2+ = reposts of reposts.

Network Node Role Distribution Over Time

What types of nodes are engaging with us? Broadcasters have asymmetric reach. Peers are balanced. Listeners consume but rarely amplify. Bridges connect otherwise separate clusters.

Topic Diffusion Heatmap: Which ideas spread furthest?

Network events per topic domain. Shows which issue areas generate the most propagation activity — not just engagement on the original post, but onward spread through the graph.

Network architecture: Every repost and quote tweet is logged as a network event. Actors are identified by one-way SHA256 hash only — the graph topology is captured without any individual being identifiable. Influence tier and node role are computed purely from public follower/following counts.

🕸 Open Interactive Network Map →

Force-directed graph — hover nodes for details, filter by topic or influence tier, trace propagation paths

4.4 — Top Performing Topics

Domain	Category	Subcategory	Posts	Avg Engagement	Best Tone
War and Peace	Active Conflicts	Gaza / Palestine	14	18.4	passionate
Economy	Housing	Rental Crisis	18	15.1	wry
Environment	Fossil Fuels	Greenwash	11	13.7	sardonic
Social Policy	Disability	NDIS	9	10.2	earnest
Domestic Politics	Political Parties	Greens	12	9.8	hopeful
Indigenous Affairs	Sovereignty	Treaty	7	8.9	earnest

4.4 — Optimal Posting Times

Window	Avg Engagement	Impression Rate	Recommendation
07:00 – 09:00 (Morning commute)	12.4	High	Priority
20:00 – 22:00 (Prime evening)	11.8	Very High	Priority
17:00 – 19:00 (Evening commute)	9.2	High	Good
12:00 – 13:00 (Lunch)	8.7	Moderate	Good
09:00 – 12:00 (Mid-morning)	5.4	Low	Moderate
22:00 – 01:00 (Late night)	4.1	Low	Low

4.5 — Audience Panel Profile (Week 4)

Gender Distribution

Female

58%

Male

33%

Nonbinary

Unknown

* Gender inferred from first name + bio pronouns. All imputed — tagged as estimated.

Age Range Distribution

25–34

40%

35–49

28%

18–24

16%

50+

Unknown

* Age inferred from bio signals + account age heuristics. Low-confidence estimates.

Top Issue Interests

Climate

72%

Anti-war

61%

Housing

54%

Disability

38%

Indigenous

31%

Profession Signals

Activist

44%

Academic

28%

Healthcare

18%

Journalist

14%

Government

10%

4.6 — Sample Tweet Performance Cards

Sammy S. @TerryS357

7:42am · Tue 22 Apr

The housing crisis isn't a supply problem. It's a political will problem. We know what works. We just keep choosing whose interests to protect instead. abc.net.au/news/…

wry critical Economy → Housing → Rental Crisis reframe

❤️ 47

🔁 23

💬 12

👁 4,200

Score: 93

Sammy S. @TerryS357

9:03pm · Thu 24 Apr

Every time we frame Gaza as "complicated," we're doing work for the people who want us to look away. It's not complicated. It's a siege. michaelwest.com.au/…

passionate angry War and Peace → Active Conflicts → Gaza / Palestine reframe

❤️ 89

🔁 41

💬 18

👁 7,800

Score: 189

Sammy S. @TerryS357 → replying to @Richard_Denniss

5:17pm · Mon 28 Apr

The structural part is right — but the harder question is why the electorate keeps accepting it. Loss aversion runs deep: people fear losing a bad deal more than they want a good one. That's what makes reform so hard even when the evidence is unambiguous.

measured questioning Economy → Inequality → General bridge thinker_reply

❤️ 34

🔁 8

💬 6

👁 2,100

Score: 56

Section 05

Privacy & Ethics Architecture

✅ What is protected

✓ Individual records never surfaced to operator
✓ Twitter user IDs hashed (SHA256, one-way, irreversible)
✓ EU/EEA users excluded at ingestion
✓ All inferred fields tagged is_imputed: true
✓ Confidence scores on all estimates
✓ Aggregates only — distributions, not individuals
✓ Data used for personal political communication only

⚠️ Considerations

⚠ AU Privacy Act — personal use exemption applies
⚠ If findings published: ethical review required
⚠ Pseudonymous account maintains professional separation
⚠ Account style/expertise may become identifiable over time
⚠ Twitter ToS: automation within API rate limits
⚠ AHPRA social media guidance: pseudonymity is correct choice

Section 06

How to Query the Data

Ask Albert at any time for an analytics report. The following queries are immediately available:

What you want	What to say to Albert
Full aggregate report	"Show me the Twitter metrics"
What tone is working best	"Which emotional tone is getting the most engagement?"
Best topics	"What topics are performing best on the account?"
Audience profile	"Tell me about the audience engaging with the account"
Best posting times	"When should I be posting for maximum reach?"
Top tweets	"What are the top performing tweets this week?"
Trend over time	"Show me the 30-day engagement trend"
Topic breakdown	"How is housing content performing vs climate content?"

Live data query: Run python3 /home/doronski/.openclaw/workspace/services/twitter/campaign/tweet_log.py metrics at any time for a full JSON analytics export.

AXIOMPolitical EngagementIntelligence System